Do you remember when you joined Twitter? I do!
Twitter used to be a place where we safely convened, shared research and met people. Now it’s a cesspool of scams, porn bots, and propaganda.
#myxanniversary
#fuckelon
Rumman Chowdhury
Rumman Chowdhury (b. April 1980) is a Bangladeshi American data scientist, a business founder, and former responsible artificial intelligence lead at Accenture. She was born in Rockland County, New York. She is recognized for her contributions to the field of data science. More on Wikipedia
-
ruchowdh.bsky.social
@ruchowdh
-
ruchowdh.bsky.social reposted
Yann LeCun
@ylecun
The UK Prime Minister has caught the Existential Fatalistic Risk from AI Delusion disease (EFRAID).
Let's hope he doesn't give it to other heads of state before they get the vaccine.
"An AI safety summit at Bletchley Park in November is expected to focus almost entirely on existential risks and how to negate them."
www.telegraph.co.uk/business/2023/09/23/artificial-intell… -
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
The computational biology people got here in 2010. With PCA. Not even deep learning.
Are we to conclude that principal components analysis has an internal world model?
pubmed.ncbi.nlm.nih.gov/20196748/ -
ruchowdh.bsky.social reposted
Gemma 🤖 ⚖️ ♀️
@gemmagaldon
#newyork tiene cosas fantásticas. Como pasar la tarde en la #highline filmando un documental con @ruchowdh y la noche en #betaworks hablando de #responsibleai
-
ruchowdh.bsky.social reposted
UC San Diego Alumni
@UCSDalumni
Congratulations to two UC San Diego alumni who landed a spot on @TIME's The 100 Most Influential People in #ai 2023 list! 👏
@ruchowdh received a doctorate in political science, and @alondra received an undergraduate degree in anthropology. 🔱 time.com/collections/time100-ai/ #time100ai -
ruchowdh.bsky.social reposted
Sayash Kapoor
@sayashk
Open foundation models are playing a key role in the AI ecosystem. On September 21, @RishiBommasani @percyliang @random_walker and I are organizing an online workshop on their responsible development. Hear from experts in CS, tech policy, and OSS.
RSVP: sites.google.com/view/open-foundation-models -
ruchowdh.bsky.social reposted
Lorena O'Neil
@lorenaoneil
There are some stories that you know will stay with you for a very long time, and this is one of them. Had the honor of profiling these five women who have been warning about AI risks before you'd ever heard of ChatGPT. My latest for @RollingStone www.rollingstone.com/culture/culture-features/women-warni…
-
ruchowdh.bsky.social reposted
Julia Angwin
@JuliaAngwin
If tech companies are going to release cars without seatbelts, then at least we need to know when & how they crash. Kudos to @ruchowdh for doing crash-testing at Defcon this week.
www.washingtonpost.com/technology/2023/08/08/ai-red-team-… -
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
"Any governance body like this needs to be independent. Companies should be engaged, but it should not be beholden to companies."
@BKCHarvard Responsible AI Fellow Dr. Rumman Chowdhury (@ruchowdh) spoke to @marketplace about ideal AI regulation
www.marketplace.org/shows/marketplace-tech/should-the-ai-… -
ruchowdh.bsky.social reposted
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
So my big presentation that's coming up is... @TEDTalks!
I'm super stoked to be speaking at the TEDWomen conference happening in October in Atlanta 💃
I'll be talking about all things AI ethics, and why we should care about its current harms and not its future (X-)risks. -
ruchowdh.bsky.social reposted
Seraphina Goldfarb-Tarrant @COLM26
@seraphinagt
We (@cohere) are participating in the DEFCON LLM Red Team challenge in 2 weeks !, spearheaded by @ruchowdh & @comathematician 🔥. 3 goals: 1. raise awareness of *real* risks of LLMs now 2. make our own model safer 3. eventually release the dataset to researchers! COME HACK US.
-
ruchowdh.bsky.social reposted
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
This 1979 (!) commentary on the perils of using the term 'emergent property' in ecology rings incredibly true even 40 years later - people keep using the term with regards to LLMs, each with their own intuitive definition.... confusing the hell out of everyone involved 😐
-
ruchowdh.bsky.social
@ruchowdh
Billy was the first to talk about underpaid data cleaning behind GenAI models (a story others are just now picking up). I love that he’s now focused on people solving the problem.
Billy Perrigo
@billyperrigo
The workers behind AI rarely reap its rewards. One nonprofit in India is trying to fix that.
After years of reporting on the darker side of the AI industry, I traveled to India with one question. Could its alternative model really work?
time.com/6297403/india-ai-karya-startup/ -
ruchowdh.bsky.social reposted
Billy Perrigo
@billyperrigo
The workers behind AI rarely reap its rewards. One nonprofit in India is trying to fix that.
After years of reporting on the darker side of the AI industry, I traveled to India with one question. Could its alternative model really work?
time.com/6297403/india-ai-karya-startup/ -
ruchowdh.bsky.social reposted
Billy Perrigo
@billyperrigo
Gig workers who make AI more reliable and less biased are often subjected to "unfair" working conditions, a new Oxford report says.
0 out of the 15 platforms @TowardsFairWork surveyed met even a "bare minimum" threshold of workplace fairness:
time.com/6296196/ai-data-gig-workers/ -
ruchowdh.bsky.social reposted
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
It blows my mind that the weather has been so unhinged in so many places this summer and yet we're still only discussing the existential risks of AI, and not climate change 🫠
-
ruchowdh.bsky.social
@ruchowdh
I support the White House commitments - but I was dismayed to see all of the media focus on for-profit companies as “external” actors when discussing commitments to red teaming. It’s good to have a startup ecosystem, but we can’t assume the market solves this problem.
-
ruchowdh.bsky.social reposted
WIRED
@WIRED
To find out more about our AI protectors read our feature, ‘Meet the Humans Trying to Keep Us Safe From AI’ at the link: wired.trib.al/INxFVp4
🎨: @samcannon -
ruchowdh.bsky.social reposted
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
It’s always “come on Barbie, let’s go party” and never “come on Barbie, come testify at my Senate hearing”
space sluts
@aerospacesluts
It’s always “come on barbie, let’s go party” and never “ come on barbie, let me fund your research”
-
ruchowdh.bsky.social
@ruchowdh
Ah yes let’s not ask a single person who has built or led a team on Ai governance because what could they possibly have to say.
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
Another US Senate hearing coming up, this time on principles of AI regulation.
I'm sure this highly diverse panel of witnesses will present complementary and level-headed testimonies that don't focus on existential risks 🫠
www.judiciary.senate.gov/committee-activity/hearings/over… -
ruchowdh.bsky.social
@ruchowdh
Watching as all the Ai “audit” firms become red teaming experts after yesterdays White House announcement gives me whiplash
Not only do none of you have any damn clue, you’re going to cause some real damage. -
ruchowdh.bsky.social reposted
CriticalAIhttps://read.dukeupress.edu/critical-ai/
@CriticalAI
From For those interested in how IP, labor, and AI intersect, an interesting article from Ryan Cooper (American Prospect) on WGA strike. prospect.org/labor/2023-07-20-unions-save-hollywood-fooli…
-
ruchowdh.bsky.social reposted
Leon Yin
@LeonYin
This is my last week at The Markup. It’s been a true privilege to practice and produce impactful hypothesis-driven journalism alongside a team of first-class journalists over the past 4 years. 🧵of some W's below:
-
ruchowdh.bsky.social
@ruchowdh
Glad that bot problem was solved
-
ruchowdh.bsky.social reposted
Yudhanjaya Wijeratne
@yudhanjaya
After a lot of exploration, I think I've solidified my thoughts on #ai and #aiart in creative work.
The way the world is headed, the haute bohème will come back.
The only art will be whatever cannot be made by a machine, a goalpost that will shift every six months.
Thread: -
ruchowdh.bsky.social
@ruchowdh
We are living in the aftermath of the fallacy that wealth equals intelligence, as exemplified by two of the wealthiest and most powerful men in the world engaging in literal dick measuring contests.
www.technologyreview.com/2023/07/14/1076296/mustafa-suley… -
ruchowdh.bsky.social reposted
Lilian Edwards
@lilianedwards
First they came for the fanfic writers and I said, er, hang on, doesn't your entire enterprise depend on copyright infringement ?www.nytimes.com/2023/07/15/technology/artificial-intellig…
-
ruchowdh.bsky.social reposted
Justin Hendrix
@justinhendrix
This is not the age of Silicon Valley innovation. It is the age of Silicon Valley bullshit. techpolicy.press/rescuing-the-future-from-silicon-valley/
-
ruchowdh.bsky.social reposted
Matt Davis @jiffyclub@hachyderm.io
@jiffyclub
So many great takeaways from @ruchowdh’s keynote on AI! I’ll try to thread some, I’m paraphrasing and summarizing so these are my own thoughts. #scipy2023
-
ruchowdh.bsky.social reposted
Neal Parikh
@npparikh
This is the exact reason I tried to explain to people this was a bad law and personally tried to kill it in NYC govt. You just create a market for a bunch of “audits” where no one knows what they are or who’s doing them. Think about markets when writing regulations.
Kate Kaye
@KateKayeReports
The NYC bias audit vendor search engine marketing begins:
-
ruchowdh.bsky.social reposted
Asmelash Teka Hadgu
@asmelashteka
Check out my "3 Minutes With" conversation:
restofworld.org/2023/3-minutes-with-asmelash-teka-hadgu/
Thank you Andrew (@decka227) for presenting our conversation so beautifully and Rest of World (@restofworld) for highlighting our story.
@lesan_ai @GhanaNLP @DAIRInstitute -
ruchowdh.bsky.social reposted
Abeba Birhane
this is not only a big embarrassing joke for all involved, including @AFP and @UN, it also shows active ignorance. one must have absolutely no idea of AI to come up with drivel like this
AFP News Agency
@AFP
VIDEO: A panel of AI-enabled humanoid robots take the microphone at a United Nations conference with the message: they could eventually run the world better than humans.
-
ruchowdh.bsky.social reposted
wells (wants a better world)
@WellsLucasSanto
2023 book pre-order recs!
* Against Technoableism by @ashleyshoo
* Death Glitch by @tamigraph
* Resurrecting the Black Body by @toniasutherland
* Your Face Belongs To Us by @kashhill
Links and descriptions in the replies! -
ruchowdh.bsky.social reposted
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
Listening to the many great talks at @aclmeeting and realizing how many depend on *access* to models and datasets for evaluating and understanding them.
What happens with the new generation of LLMs, which are only available via commercial APIs? How can we do science on them? -
ruchowdh.bsky.social reposted
Roger McNamee
@Moonalice
Chat GPT is optimized to fool users into believing it is human. It does not do this by being smart. It does it by imitating the most convincing language in its training set and then filling in the details with both facts and lies.
Carissa Véliz
@CarissaVeliz
"Not only could the majority of participants not tell the real tweets from the ones generated by AI, but they generally found the AI's tweets more convincing." #aiethics #llms
www.businessinsider.com/ai-generated-tweets-study-openai-… -
ruchowdh.bsky.social reposted
Prof. Dr Beth Singler.
@BVLSingler
I've been thinking for a while that some people might be more sensitive to the uncanniness or 'inbredness' of AI generated people than others - much as CGI set off people's uncanny meters and others found it fine. I played with the idea in this story too: bvlsingler.com/2020/11/24/new-short-story-uncanny-sensiti…
ruchowdh.bsky.social
@ruchowdh
You know how AI gen people have this creepy manic look about them (massive smiles) because the model is overfit on certain attributes?
Kind of like how homogenous gene pool in dogs or people leads to augmentation of features or disease?
I’ve started calling it AI inbreeding. -
ruchowdh.bsky.social
@ruchowdh
You know how AI gen people have this creepy manic look about them (massive smiles) because the model is overfit on certain attributes?
Kind of like how homogenous gene pool in dogs or people leads to augmentation of features or disease?
I’ve started calling it AI inbreeding. -
ruchowdh.bsky.social reposted
@DRJessieNYC 🇵🇸
@JessieNYC
"The whole thing was bullshit. Far from being worth trillions of $$$, the Metaverse turned out to be worth absolutely bupkus."
www.thenation.com/article/culture/metaverse-zuckerberg-pr… -
ruchowdh.bsky.social reposted
shoshana wodinsky (she/her)
@swodinsky
to anyone who thinks bot-written articles—on gizmodo or anywhere else—are meant for humans. they’re not. they’re meant for other bots
-
ruchowdh.bsky.social
@ruchowdh
Are you all really gonna make me download threads tho
-
ruchowdh.bsky.social reposted
Sayash Kapoor
@sayashk
OpenAI's ChatGPT lost its browsing feature a couple of days ago, courtesy of @random_walker's demonstration that it could output entire paywalled articles.
But Bing itself continues to serve up paywalled articles, word for word, no questions asked. -
ruchowdh.bsky.social
@ruchowdh
I love this so much. Going to add this to my pile of “things bros say to me”
Lech Mazur
@LechMazur
Indeed, I am the judge of your AI knowledge. It was a quick judgment, but no doubt accurate.
It doesn't seem like I'm the one who needs to take deep breaths here. Nice projection. -
ruchowdh.bsky.social reposted
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
Hey @samcharrington, did you know *a lot* of @twimlai content is in LAION? 🙃
Spawning
@spawning_
New feature: haveibeentrained.com just got a *lot* more powerful with caption search 🔎
If you search your name, many more results will appear than before!
If you are present in popular AI image datasets, this search will find you, and you can opt out in a few clicks :) -
ruchowdh.bsky.social reposted
Nitasha Tiku
@nitashatiku
New from me. Wondering how “existential risks” around AI causing human "extinction" became mainstream? It’s part of a tech-billionaire backed effort that relied on colleges to grow—both as a pathway to prestige & recruiting ground for idealistic talent www.washingtonpost.com/technology/2023/07/05/ai-apocalyps…
-
ruchowdh.bsky.social reposted
Shannon Liss-Riordan
@SLissRiordan
But first - we need to compel Twitter to arbitrate the nearly 2,000 claims we have brought. Apparently it is now "sinking in" to Elon how expensive his gambit was to force employees to bring claims against Twitter one-by-one. www.theregister.com/2023/07/05/twitter_arbitrations_lawsuit/
-
ruchowdh.bsky.social
@ruchowdh
Latest OpenAI blog post
-
ruchowdh.bsky.social reposted
Xiao Ma
@infoxiao
Well don’t I have news for you @pmarca @benthompson.
Our recent work shows moral reasoning is among the worst performing tasks for language models. I have countless examples showing how discussing moral scenarios goes horribly wrong.
arxiv.org/abs/2306.14308 -
ruchowdh.bsky.social reposted
neil turkewitz
@neilturkewitz
“The plaintiffs state that OpenAI is now worth almost $30 billion, & yet, ‘the individuals & co’s that produced the data it’s scraping haven’t been compensated.’ This action ‘seeks to change that, & in the process, protect the privacy rights of millions.’” www.thefashionlaw.com/openai-sued-over-unprecedented-data…
-
ruchowdh.bsky.social
@ruchowdh
On all ends, AI is being used to consolidate power. Imagine industry giants churning out branded content for decades longer than they usually would; pushing out the next generation of up and comers
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
So... I guess even JLo is going on the generative ("Jen" lol) AI train?
voyagewithjen.ai
Straight out of Black Mirror 😆 -
ruchowdh.bsky.social reposted
GMG Union
@gmgunion
Our statement on G/O Media’s plan to implement AI content, just days after laying off newsroom members
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Hello NeurIPS community,
the 🚨 Call for Ethics Reviewers 🚨 is still open!
1/ We are still looking for -more- second-round reviewers, esp. in 1) Human rights and inappropriate potential applications, and 2) Security and privacy! Please sign up through the Google Form below.
Jiahao Chen @NeurIPS
@acidflask
🚨 Calls for Ethics Reviewers 🚨
@NeurIPSConf is seeking second-round reviewers focusing on ethics and broader impact. Your service to the premier AI/ML research conference will be greatly appreciated!
Call: neurips.cc/Conferences/2023/CallForEthicsReviewers
Sign up: tiny.cc/neurips2023ethicssignup -
ruchowdh.bsky.social reposted
Edwin Lane
@EdwinLane
A new season of Tech Tonic from the @FinancialTimes! Have we reached peak social media? Episode 1 on this website, and whether @elonmusk is going to save it or destroy it. Featuring @ruchowdh @rabble, the FT's own @MsHannahMurphy, hosted by @ElaineDMoore: link.chtbl.com/TechTonic
-
ruchowdh.bsky.social
@ruchowdh
Not only is “explainable” Ai difficult it’s also insufficient. Explainability means nothing if power dynamics are not addressed.
Will Henshall
@henshall_will
My first piece for TIME: Chuck Schumer Wants AI to Be Explainable. It’s Harder Than It Sounds.
time.com/6289953/schumer-ai-regulation-explainability/ -
ruchowdh.bsky.social reposted
Tech Policy Press
@techpolicypress
Transcript: House Hearing on AI- Advancing Innovation in the National Interest
Witnesses:
-Rand CEO Jason Matheny
-Lux Capital's Shahin @Farshchi
-@huggingface CEO @ClementDelangue
-Harvard's Rumman Chowdhury (@ruchowdh)
-@CSETGeorgetown's Dewey Murdick
techpolicy.press/transcript-house-hearing-on-ai-advancing… -
ruchowdh.bsky.social reposted
AlexandraReeveGivens
@AlexReeveGivens
“Governance is a spectrum ranging from norms & codes of conduct, to standards & benchmarks, to oversight & regulation…[T]he US needs to focus on significant investment in all aspects of [#ai] governance.” Thoughtful words from @ruchowdh today science.house.gov/2023/6/artificial-intelligence-advancin…
-
ruchowdh.bsky.social reposted
Kellee Wicker
@kelleewicker
"Brakes help you drive faster" - @ruchowdh laid out a great argument for why trust is key for innovation and gov't action and regulation is foundational for that trust in her testimony before House Science today, watch more on the livestream: science.house.gov/2023/6/artificial-intelligence-advancin…
-
ruchowdh.bsky.social reposted
Kevin Bankston
@KevinBankston
Gotta love HCSST committee. A refreshingly substantive and thoughtful hearing tackling some really nuanced questions. Particularly appreciated @ruchowdh on need for field-building investment in AI accountability experts, and @ClementDelangue on the many benefits of open source AI
Ted Lieu
@tedlieu
Our Science, Space and Technology Committee is holding a fascinating hearing on AI. Watch now.
#thursdayvibes -
ruchowdh.bsky.social
@ruchowdh
My proudest moment was getting the term “Taco Tuesday” on the congressional record.
-
ruchowdh.bsky.social
@ruchowdh
Hugging face-ing with @ClementDelangue
-
ruchowdh.bsky.social reposted
Kelvin Yu
@kelvinotcelsius
Tune in at 10am ET today to hear our esteemed witnesses testify on how to advance AI in our national interest @JasonGMatheny @Farshchi @ClementDelangue @DeweyAM @ruchowdh
science.house.gov/2023/6/artificial-intelligence-advancin… -
ruchowdh.bsky.social
@ruchowdh
Honestly I can’t even compose a snarky tweet because this is just so fucking bad.
- dataset of hotness based on 60 peoples input 🙅♀️ (tldr “ladies look cuter!”)
- using facial morphology to determine “masculinity” 🙊
- the sheer whiteness of this all 😬
www.nature.com/articles/s41598-023-31796-1 -
ruchowdh.bsky.social reposted
Break Through Tech DC at UMD
@BrkThroughUMD
Interest in #artificialintelligence has exploded.
Hear about responsible AI from @BKCHarvard's @ruchowdh and the intersection of tech with gender/race from @sbuartofficial's @stephdink in this creative @nytimes piece by @phoebelett and @jashkenas. www.nytimes.com/interactive/2023/06/01/opinion/ai-technol… -
ruchowdh.bsky.social reposted
Dr. Joy Buolamwini
@jovialjoy
I am looking forward to this meeting with @POTUS on the dangers of AI and what we can do to prevent harms already impacting everyday people seeking mortgages and housing, in need of medical treatment, encountering workplace surveillance, & more @AJLUnited
-
ruchowdh.bsky.social
@ruchowdh
Unsurprisingly for bird app -
GLAAD
@glaad
This year, GLAAD’s Social Media Safety Index found that all five major social media platforms – Facebook, Instagram, TikTok, YouTube and Twitter – received failing scores on the SMSI Platform Scorecard for the second consecutive year: glaad.org/smsi/lgbtq-social-media-safety-program/
-
ruchowdh.bsky.social reposted
Dorothea Baur (Dr.)
@DorotheaBaur
"when we ask whether 'AI will destroy/save us,' we erase the fact that human beings create and deploy #ai in the first place".
But "(h)umanity isn’t an inefficiency to be optimized away but something to be carefully...nurtured"
Great piece by @ruchowdh! thehill.com/opinion/technology/4047323-artificial-intelli… -
ruchowdh.bsky.social reposted
Boston Public Radio
@BosPublicRadio
Responsible AI fellow @ruchowdh and @WIRED senior writer @willknight join us now to discuss the future of artificial intelligence and where it could fit into our everyday lives. Listen now: www.wgbh.org/news/boston-public-radio
-
ruchowdh.bsky.social reposted
Sheera Frenkel
@sheeraf
Misinfo experts are being inundated with subpoenas, lawsuits, and legal letters which have slowed their work to a crawl ahead of the 2024 elections. (The effort is being organized by some of the #stopthesteal crowd who spread misinfo in 2020).
www.nytimes.com/2023/06/19/business/gop-disinformation-re… -
ruchowdh.bsky.social reposted
Rachel Metz
@rachelmetz
It’s Saturday, which means it’s time for a reminder that papers published to the arxiv are *not* peer reviewed and should be approached with caution (if not skepticism), particularly if you’re a journalist (or anyone interested in sharing research with a large audience).
-
ruchowdh.bsky.social reposted
DFRLab - @dfrlab.bsky.social
@DFRLab
WEDNESDAY - Join @DFRLab for the launch of our Task Force for a Trustworthy Web ft. @rightsduff, @Rosejackson3, @_JustinDavis, @ruchowdh, @AlexReeveGivens, @micheleflournoy, Lauren Buitta, @brianfishman as they discuss generative AI and national security.
www.atlanticcouncil.org/event/scaling-trust-on-the-web/ -
ruchowdh.bsky.social
@ruchowdh
I’ll be testifying to Congress next week on the need for Trustworthy AI - I’ll post my testimony here (as soon as I’m done writing it 😂)
science.house.gov/2023/6/artificial-intelligence-advancin… -
ruchowdh.bsky.social reposted
Martin Tisné
@martintisne
"A technocentric world is based on the premise that humanity is flawed and technology will save us" 👏 @ruchowdh & @SEHendrickson thehill.com/opinion/technology/4047323-artificial-intelli…
-
ruchowdh.bsky.social reposted
neil turkewitz
@neilturkewitz
💯 We need to wake them up from their tech-exceptionalist slumber. As @ruchowdh & @SEHendrickson wrote—“our worst-case scenario is a technocentric world where the blind pursuit of AI growth & optimization outweighs the imperative for human flourishing.”
thehill.com/opinion/technology/4047323-artificial-intelli… -
ruchowdh.bsky.social
@ruchowdh
My latest, for @thehill , with BKC Director Sue Hendrickson -
“The worst-case scenario about AI isn’t about AI at all. It’s about humans making active decisions to pursue technological growth at all costs.”
thehill.com/opinion/technology/4047323-artificial-intelli… -
ruchowdh.bsky.social reposted
Reuters Institute
@risj_oxford
Trust in the news has fallen across markets by a further 2 points in the last year, reversing in many countries the gains made in the pandemic
📊On average, 40% say they trust most news most of the time. 🇫🇮 has the highest levels of trust (69%) while 🇬🇷 (19%), the lowest -
ruchowdh.bsky.social reposted
Nanjala Nyabola
@Nanjala1
I'm finally over my jetlag and I can speak a little bit about our experiences last week in Panama and Costa Rica as part of RightsCon. I want to caveat all this by saying I don't work for Access Now although I do serve on the board. Everything I'm saying here is on my own behalf.
-
ruchowdh.bsky.social reposted
Jonathan Shedler
@JonathanShedler
This study suggests that people who publicly shame & dogpile on social media are not motivated by moral outrage or a desire to do good, but by sadism
They just get off on the cruelty
papers.ssrn.com/sol3/papers.cfm?abstract_id=4359608 -
ruchowdh.bsky.social
@ruchowdh
The ahistorical nature of tech discovering democratic processes is simultaneously funny and sad to me, a political scientist. Rivers of ink have been spilled on the efficacy of participatory processes. The short is, they’re hard and they don’t scale. 1/
www.noemamag.com/participation-at-scale-can-repair-the-pu…
ruchowdh.bsky.social
@ruchowdh
The main basic flawed premises here are
1/ majority rule is always best
2/ people are or want to be informed about tech regularly
3/ this decisioning can move fast enough to make tech-speed decisions
4/ this method can be applied to any size of population
ruchowdh.bsky.social
@ruchowdh
I wonder if people have asked themselves why no major democracy is organized like this?
This really isn’t me being facetious- I want to know what empirical work has been done to identify the flaws of analog deliberative processes and how technology has solved those problems. -
ruchowdh.bsky.social reposted
LLM Security
@llm_sec
* People ask LLMs to write code
* LLMs recommend imports that don't actually exist
* Attackers work out what these imports' names are, and create & upload them with malicious payloads
* People using LLM-written code then auto-add malware themselves
vulcan.io/blog/ai-hallucinations-package-risk -
ruchowdh.bsky.social reposted
Grace
@gracecamille_
Some of u are about to b mad
Nooruddean
@BeardedGenius
Who's the definition of a 'stupid person's idea of a clever person'
-
ruchowdh.bsky.social
@ruchowdh
.@rachelcoldicutt with the realest take on our godfathers of AI
rachelcoldicutt.medium.com/on-understanding-power-and-tec…
ruchowdh.bsky.social
@ruchowdh
My life - “By way of making conversation, the Very Eminent Computer Scientist asked me what I “do”. This is an oddly complicated question at the best of times, because I’m a chronic overthinker, don’t have a convenient Richard Scarry job title, and I do a few different things.”
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
🚨 Calls for Ethics Reviewers 🚨
@NeurIPSConf is seeking second-round reviewers focusing on ethics and broader impact. Your service to the premier AI/ML research conference will be greatly appreciated!
Call: neurips.cc/Conferences/2023/CallForEthicsReviewers
Sign up: tiny.cc/neurips2023ethicssignup -
ruchowdh.bsky.social reposted
Rest of World
@restofworld
The lack of digital literacy in India makes job seekers particularly vulnerable to being conned. “People aren’t really able to tell the difference between what’s genuine outreach versus what is likely to be a scam”
restofworld.org/2023/india-job-crisis-scams-spike/ -
ruchowdh.bsky.social
@ruchowdh
So Hinton and Bengio are the “godfathers” and Altman is a “father” - what’s the title for the women who are leading the charge to protect humanity?
Or is that labor just expected?
Guardian Tech
@guardiantech
‘What should the limits be?’ The father of ChatGPT on whether AI will save humanity – or destroy it www.theguardian.com/technology/2023/jun/07/what-should-th…
-
ruchowdh.bsky.social reposted
Emily Ekins
@emilyekins
NEW CATO SURVEY: Nearly a third of Gen Z support the government installing surveillance cameras in every household to prevent crime and abuse.
Opposition to Orwelian-government monitoring has been assumed to be both overwhelming and obvious. But is it?
A #catopolls [thread]… -
ruchowdh.bsky.social reposted
Duolingo English Test
@DuolingoENTest
🚀 The future
@esakismith led a panel of experts in test assessment (our vey own @AlinaVdav), literature, plagiarism software, & machine learning ethics (@ruchowdh) to discuss their thoughts on the future of generative A.I., while asking: "What exactly is at stake?" -
ruchowdh.bsky.social reposted
Jonathan Cheng
@JChengWSJ
Artificial intelligence is making its way into daily Chinese life—including a surge of AI-driven fraud. In recent weeks, police across China have warned about criminals employing AI software to impersonate friends of their targets.
@shenlulushen
www.wsj.com/articles/china-cracks-down-on-surge-in-ai-dri… -
ruchowdh.bsky.social reposted
Karla Monterroso
@karlitaliliana
It is wild to me how many people gave Jack the benefit of the doubt, over and over again.
NR_Garrett
@NR_Garrett
Everything that’s wrong with the world in 3 tweets
-
ruchowdh.bsky.social reposted
Block Party
@blockpartyapp_
Privacy Party is a browser extension that reduces your risk of harassment, cyberstalking, impersonation, fraud, spam, and other online attacks with automated privacy playbooks for your social media accounts.
-
ruchowdh.bsky.social reposted
odbol
@odbol
"The things we should be asking are what are the things that are wrong with society today and how might that be reflected in the systems we are building?" – @ruchowdh www.nytimes.com/interactive/2023/06/01/opinion/ai-technol…
-
ruchowdh.bsky.social reposted
A/Prof. Dana McKay
@DanaChatter
This is a term I needed to find: 'moral outsourcing' is when technologists ascribe the harm their technologies cause to the technologies themselves (see 'biased AI'), and thus evade responsibility for it. THANK YOU @ruchowdh www.theguardian.com/technology/2023/may/29/rumman-chowdhu…
-
ruchowdh.bsky.social reposted
New York Times Opinion
@nytopinion
Dr. Rumman Chowdhury (@ruchowdh) was the director of machine learning transparency, ethics and accountability at Twitter. www.nytimes.com/interactive/2023/06/01/opinion/ai-technol…
-
ruchowdh.bsky.social reposted
New York Times Opinion
@nytopinion
What should we expect as artificial intelligence improves?
In a special project from @phoebelett and @jashkenas, people who use A.I. discuss its perils and promise, with their arguments illustrated by A.I.-generated video. www.nytimes.com/interactive/2023/06/01/opinion/ai-technol… -
ruchowdh.bsky.social reposted
Sven Cattell
@comathematician
Deep Learning will never be solved. This is a 2D slice of a simple MNIST network with 64 hidden layers. Each color represents a unique linear region.
The messed up thing is each training point is in it's own unique region, and TL;DR: this makes security hard.
Rob Miles
@robertskmiles
Their results are bizarre and inhuman. @NeelNanda5 trained a tiny transformer to do addition, then spent weeks figuring out what it was doing - one of the only times in history someone has understood how a transformer works.
This is the algorithm it created. To *add two numbers*! -
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
Must read profile by @whatpaula in @guardian of @BKCHarvard Responsible AI Fellow Dr. Rumman Chowdhury @ruchowdh
www.theguardian.com/technology/2023/may/29/rumman-chowdhu… -
ruchowdh.bsky.social reposted
Chloe Xiang 向舒锦
@chloexiang
Today, the NEDA took the chatbot down after it produced a number of harmful responses, such as encouraging unhealthy eating habits rather than helping someone with an eating disorder. www.vice.com/en/article/qjvk97/eating-disorder-helpline-d…
-
ruchowdh.bsky.social reposted
fran
@galacticidiots
Jane Austen’s works are timeless classics because she knew the real horrors in life are having to listen to men who think they’re better than you and receiving unannounced visitors.
-
ruchowdh.bsky.social reposted
paula aceves
@whatpaula
In light of all the “AI might kill us but we promise we’re trying to stop it” news, @ruchowdh reminds us that behind every machine, there is a person. www.theguardian.com/technology/2023/may/29/rumman-chowdhu…
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Working with the IAEA really puts into perspective how immature the AI x-risk discourse is compared to preventing an actual nuclear power plant in Zaporizhzhiya blowing up. We have so much to learn about risk management, emergency response and crisis communications from the pros
-
ruchowdh.bsky.social
@ruchowdh
You know if they were so concerned they could just NOT build the technology.
Dan Hendrycks
@hendrycks
We just put out a statement:
“Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.”
Signatories include Hinton, Bengio, Altman, Hassabis, Song, etc.
safe.ai/statement-on-ai-risk
🧵 (1/6)
ruchowdh.bsky.social
@ruchowdh
I’m glad @whatpaula revived my old talk on moral outsourcing, because this is a clear example. The invocation of “AI” by the CEOs of the companies building it is a classic example
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Humbled to be working with @iaeaorg @ruchowdh in Vienna on the uses of AI for crisis communications and disinformation risks
-
ruchowdh.bsky.social reposted
Tech Policy Press
@techpolicypress
Last week, the Knight Foundation and Georgetown announced the creation of the Knight-Georgetown Institute, “a central hub for the growing network of scholarship that seeks to shape how technology is used to produce, disseminate and access information.”
techpolicy.press/new-institute-at-georgetown-to-translate… -
ruchowdh.bsky.social
@ruchowdh
Day 2 at the UN in Vienna. The #iaea has become an exemplar of global governance.
Let’s test that - how will a global agency with non enforcement powers address the issue of GenAI based misinfo in nuclear crisis?
I’m working w/ a wide ranging group of experts to address this. -
ruchowdh.bsky.social
@ruchowdh
Deeper than this - some NEED to think this because we are taught our self worth comes from superiority. It’s hard to be a non competitive person at a top tier school.
-
ruchowdh.bsky.social
@ruchowdh
Thank you to @guardian reporter Paula Aceves for this profile - the part she left out is she was patient enough to sit through my jet lagged chaos over lunch 😀 I appreciate her distillation of both my frustration with the field but also my optimism
www.theguardian.com/technology/2023/may/29/rumman-chowdhu… -
ruchowdh.bsky.social reposted
Scroll.in
@scroll_in
Replug | "India has developed a very unique deep state. Deep state usually means the secret parts of the state machinery, which are at work trying to subvert existing governance structures."
Interview with @josyjosephkj | scroll.in/article/1006730/ -
ruchowdh.bsky.social
@ruchowdh
Shocked pikachu
BBC News Technology
@BBCTech
Twitter pulls out of voluntary EU disinformation code www.bbc.com/news/world-europe-65733969
-
ruchowdh.bsky.social reposted
Sven Cattell
@comathematician
We did a thing. @austincarson @ruchowdh
www.nbcnews.com/now/video/how-ai-hacking-event-will-help-… -
ruchowdh.bsky.social
@ruchowdh
What a strange headline by NYT for an article that details how expert academics, civil society and companies don’t like the NYC hiring law
www.nytimes.com/2023/05/25/technology/ai-hiring-law-new-y… -
ruchowdh.bsky.social reposted
ruchowdh.bsky.social
@ruchowdh
Eep! We were on NBC News talking about the generative AI red teaming challenge @aivillage_dc!
www.nbcnews.com/now/video/how-ai-hacking-event-will-help-… -
ruchowdh.bsky.social
@ruchowdh
Eep! We were on NBC News talking about the generative AI red teaming challenge @aivillage_dc!
www.nbcnews.com/now/video/how-ai-hacking-event-will-help-… -
ruchowdh.bsky.social reposted
Lilian Edwards
@lilianedwards
My Turing lecture on generative AI regulation and panel with wonderful @wsisaac @boom_lawyered and compere @adrian_weller is now up youtu.be/bMxBTYqDyOI - sorry very bad sound near beginning.
Tomorrow, latest iteration at LSE! www.imperial.ac.uk/news/244993/imperial-partners-with-lse… -
ruchowdh.bsky.social reposted
The PIE News
@ThePIENews
In a @DuolingoENTest #detcon23 panel facilitated by Anna Esaki-Smith, experts from Duolingo English Test, Great Books, Turnitin, and Parity Consulting discuss the challenges and opportunities of AI in higher education. #ai #highereducation #intled #technologysolutions
-
ruchowdh.bsky.social reposted
WIRED
@WIRED
This week in our first ever crossover episode on 'Have a Nice Future', @glichfield and @LaurenGoode sit down with @senari cohost of The Pitchfork Review podcast, to discuss how AI is changing music. wired.trib.al/NgmoEAH
-
ruchowdh.bsky.social reposted
Martin Bauer
@martinmbauer
Many of you aren’t aware of what is going on at UK universities
A spiral of low pay, pension cuts and casual contracts have plunged the nation into strike
Students won’t graduate. Some of the most talented people will leave. Universities will lose accreditation
1/4 -
ruchowdh.bsky.social reposted
Matt Cagle
@Matt_Cagle
This violates Twitter’s longstanding policy which bans developers from creating surveillance products with user info. The company booted federal fusion centers in 2016 for similar violations. What has changed? 🤔www.aclu.org/news/privacy-technology/twitter-cuts-fusion-…
-
ruchowdh.bsky.social
@ruchowdh
Just a reminder that this is coming up in July!
ruchowdh.bsky.social
@ruchowdh
CFP: Yesterday was a big day in Generative AI. What follows the is the convo about AGI and sentience. But what IS sentience?
I’m guest editing a special edition of JoSoCo to uncover whether AGI is reality or a red herring: 1/
www.sciopen.com/journal/message_news/get_by_id?id=1634013… -
ruchowdh.bsky.social reposted
Rachel Tobac
@RachelTobac
Here’s how I used AI to clone a 60 Minutes correspondent’s voice to trick a colleague into handing over her passport number. I cloned Sharyn’s voice then manipulated the caller ID to show Sharyn’s name with a spoofing tool.
The hack took 5 minutes total for me to steal the info.
60 Minutes
@60Minutes
60 Minutes hired an ethical hacker to show how easy it is to be scammed. She conned our unsuspecting colleague using artificial intelligence. www.cbsnews.com/news/how-con-artists-use-ai-apps-to-steal…
This media couldn't be displayed. See it on the original post. -
ruchowdh.bsky.social
@ruchowdh
Ok so at this point I’ve had multiple interactions where it’s our various OOOs talking to each other.
Can we all just admit we are overloaded 😞 -
ruchowdh.bsky.social reposted
Carissa Véliz
@CarissaVeliz
“I haven’t worked in the last four [years] because I’m too expensive,” Mr. Hume said. “And that’s mostly because #netflix broke the model. I think they put all the money into production in the streaming wars, and they took it away from #writers.”
www.nytimes.com/2023/05/11/business/media/netflix-writers… -
ruchowdh.bsky.social reposted
Queer In AI (QAI)
@QueerinAI
Our organizers @multilingual_s and @willie_agnew talked with Nature about Queer in AI, which also features @QuickResolute and @DeepIndaba. Have a look at the article: www.nature.com/articles/d41586-023-01689-4
-
ruchowdh.bsky.social
@ruchowdh
“Accounting” is actually wildly popular among students and recent grads idk what they’re saying
The Wall Street Journal
@WSJ
Academics, recruiters and college graduates say low starting pay and onerous testing are damping demand for accounting degrees and discouraging graduates from pursuing accounting careers www.wsj.com/articles/why-graduates-arent-hot-on-accountin…
-
ruchowdh.bsky.social reposted
Mark Riedl
@mark_riedl
*opinion* Altman would rather talk about regulation of sci-fi scenarios and not about the real world consequences of how AI can be used right now to do harm. Things like protection from discrimination, fairness, privacy, ability to receive remedy... 1/3
-
ruchowdh.bsky.social
@ruchowdh
Ozempic is the new AI
Sarah Zhang
@sarahzhang
All these behaviors likely have overlapping pathways in the brain.
So, Ozempic doesn't just affect the gut. It's probably acting directly on the brain.
www.theatlantic.com/health/archive/2023/05/ozempic-addict… -
ruchowdh.bsky.social
@ruchowdh
Well this is gross.
Lee Fang
@lhfang
Buzzfeed Inc. (owners of Complex, HuffPost, Tasty, etc) tells investors that AI will generate content, headlines, "infinite quizzes" & develop Black, Asian, Latino identity-based content to help corporate brands tap an "authentic voice" to sell products investors.buzzfeed.com/static-files/b5474867-1907-415c-bb…
-
ruchowdh.bsky.social reposted
Lee Fang
@lhfang
Buzzfeed Inc. (owners of Complex, HuffPost, Tasty, etc) tells investors that AI will generate content, headlines, "infinite quizzes" & develop Black, Asian, Latino identity-based content to help corporate brands tap an "authentic voice" to sell products investors.buzzfeed.com/static-files/b5474867-1907-415c-bb…
-
ruchowdh.bsky.social
@ruchowdh
We've had so much interest in our @aivillage_dc LLM Red Teaming challenge! We want to be as inclusive of the community as possible so if you are interested in volunteering (IRL or virtual) pls sign up here. You don’t have to be an ml security expert!
docs.google.com/forms/d/1VzLzZxk0OB_6nFqw07IqBJSB46j8I8_K… -
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
MAY 22: Join us for a virtual conversation with @JuliaAngwin, @BrandonSilverm, and @BKCHarvard Responsible AI fellow Dr. Rumman Chowdhury (@ruchowdh) on enabling effective and accountable technical oversight of generative AI.
cyber.harvard.edu/events/enabling-accountable-technical-o… -
ruchowdh.bsky.social
@ruchowdh
The third in our series on governing generative AI - how do we arrive at meaningful technical accountability?
Berkman Klein Center for Internet & Society
@BKCHarvard
MAY 22: Join us for a virtual conversation with @JuliaAngwin, @BrandonSilverm, and @BKCHarvard Responsible AI fellow Dr. Rumman Chowdhury (@ruchowdh) on enabling effective and accountable technical oversight of generative AI.
cyber.harvard.edu/events/enabling-accountable-technical-o… -
ruchowdh.bsky.social reposted
Carissa Véliz
@CarissaVeliz
Given the lack of regulation and the state of #aiethics, this is a great initiative. "Chowdhury (@ruchowdh), co-founder of #ai accountability nonprofit Humane Intelligence, said it’s not just about finding flaws but about figuring out ways to fix them." www.nbcnews.com/tech/tech-news/mass-event-will-let-hacker…
-
ruchowdh.bsky.social
@ruchowdh
One aspect of this licensing debate that is under discussed is the role of culture - in this case the culture of data science/ML/AI 🧵
clem 🤗
@ClementDelangue
Requiring a license to train models would be like requiring a license to write code. IMO, it would further concentrate power in the hands of a few & drastically slow down progress, fairness & transparency.
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
To put a finer point- we have no centralized accreditation body; we learn in a decentralized and collaborative fashion. I can also see how that is completely incongruent with how regulation or licensing would work, even if at face I understand the appeal of licensing
ruchowdh.bsky.social
@ruchowdh
Tl dr - many of us agree that there needs to be standards of appropriate use. IMO at the core of resolving this isn’t creating a group (that won’t be seen as legitimate) and forcing it, it comes from a careful recalibration and appreciation of our culture, community and norms
-
ruchowdh.bsky.social reposted
Daron Acemoglu
@DAcemogluMIT
My book, co-authored with @baselinescene, Power and Progress: Our Thousand-Year Struggle over Technology and Prosperity is out today (shapingwork.mit.edu/power-and-progress/).
Here is a thread to explain why we wrote it (and why we are excited to share the ideas that are basis of the book). -
ruchowdh.bsky.social
@ruchowdh
MAKE THIS MAKE SENSE
Melanie Mitchell
@MelMitchell1
Yeah, the Senate hearings on AI regulation are going great.
(from www.washingtonpost.com/technology/2023/05/16/ai-congressi…) -
ruchowdh.bsky.social
@ruchowdh
I did a thing
-
ruchowdh.bsky.social
@ruchowdh
Cut out the middleman and just date ChatGPT
Anna Tong
@annatonger
.Stanford student @bryanhpchiang created rizzGpt, a monocle you wear that displays chatGPT output to tell you what to say in case you get tongue tied on a date. He calls it “CaaS” or charisma-as-a-service.
-
ruchowdh.bsky.social reposted
Roger McNamee
@Moonalice
No one has less credibility than Schmidt. He has been exactly saying this for 20 years, as his co & Big Tech undermined pub. health, democracy, and pub. safety. AI is not complicated. Training sets scraped from web are dangerous. Ad monetization will increase harm. Regulate now.
Alex Thompson
@AlexThomp
Eric Schmidt,an active player in how Biden admin handles AI, says companies shud be regulating themselves rather than the govt.
“There’s no way a non-industry person can understand what’s possible….theirs no one in the government who can get it right”This media couldn't be displayed. See it on the original post. -
ruchowdh.bsky.social reposted
Cher Scarlett 🌌
@cherthedev
So.
Child Protective Services Screening has shifted to using AI (automated decision making) to determine whether or not the risk is worth investigating.
Why? There's too many calls and not enough child welfare workers.
We're solving the problem of resource starvation with AI. -
ruchowdh.bsky.social reposted
Şerife (Sherry) Wong
@sherrying
Notable from a different religious LLM site, HadithGPT "After feedback from the community, I decided to shut it down and consult with scholars to find a better use for this technology. Thank you all for your support!" h/t @DrSyedMustafaA1 restofworld.org/2023/chatgpt-religious-chatbots-india-git…
-
ruchowdh.bsky.social reposted
Jack Stilgoe
@Jackstilgoe
The editor of Science reminds his readers that scientists are human, but forgets to mention that it took an anthropologist (Amy Moran-Thomas) to start correcting the science on pulse oximeters www.science.org/content/blog-post/it-matters-who-does-sci…
-
ruchowdh.bsky.social reposted
Kim Brunhuber
@kimbrunhuber
"There's potential real accountability that is at stake here" with new AI rules for the EU, which could make their way across the pond due to the so-called Brussels effect. @sarahchander with @edri walks me through draft rules for ChatGPT.
-
ruchowdh.bsky.social reposted
M. Alex O. Vasilescu
@AlexTensor
Attn: @MIT, @MITCSAIL, @MITMath @Medialab alumni
Multiple generations of students have taken Gilbert Strang's linear algebra course.
Prof. Gilbert Strang's last lecture is ... Monday!
Watch it via live stream: grinfeld.org/strang/
MIT Mathematics
@MITMath
Prof. Gil Strang's last lecture is ... Monday! Watch it via live stream: grinfeld.org/strang/
-
ruchowdh.bsky.social reposted
No Context Russian Doll
@russiandollnc
-
ruchowdh.bsky.social reposted
MMitchell
@mmitchell_ai
I iz h@x0r and you can b2.
ruchowdh.bsky.social
@ruchowdh
This is my skeptical face - but also, great piece on our genAI red teaming exercise at DEFCON AI village
apnews.com/article/hacking-jailbreaking-chatgpt-bing-defc… -
ruchowdh.bsky.social reposted
Isabel Pedersen
@isabel_pedersen
“We need thousands of people,” Rumman Chowdhury @ruchowdh DEF CON hacker convention “We need a lot of people with a wide range of lived experiences, subject matter expertise & backgrounds hacking at these models, trying to find problems that can be fixed”apnews.com/article/hacking-jailbreaking-chatgpt-bing-defc…
-
ruchowdh.bsky.social
@ruchowdh
This is my skeptical face - but also, great piece on our genAI red teaming exercise at DEFCON AI village
apnews.com/article/hacking-jailbreaking-chatgpt-bing-defc… -
ruchowdh.bsky.social
@ruchowdh
My experience in the @defcon community has been minimal, and limited mostly to @aivillage_dc but it’s so wonderful to be in a collaborative and respectful group working towards a positive shared goal. Cc @comathematician @austincarson
-
ruchowdh.bsky.social reposted
Reclaim Your Face
@ReclaimYourFace
1/7 🗓️Tomorrow, @Europarl_EN has an opportunity to make history.
So much of what we have been fighting for for years as #reclaimyourfacewill be on the line.
Read on to understand what's happening, and why it matters, in the fight against biometric #masssurveillance (BMS)🧵 -
ruchowdh.bsky.social reposted
AlgorithmWatch
@algorithmwatch
Finally, Members of @Europarl_EN are voting on the #aiact this week.
We desperately need transparency about the context in which #algorithms are used to understand how they impact us.
We urge MEPs to vote YES to more stringent transparency obligations in the upcoming vote. ✊ -
ruchowdh.bsky.social reposted
𝐂+𝐌 𝐒𝐢𝐥𝐯𝐞𝐫 𝐂𝐞𝐧𝐭𝐞𝐫 @𝐍𝐘𝐔
@CMSCenterNYU
🗨️Join us on May 18th at 11 AM EDT for a seminar with Dr. Rumman Chowdhury (@ruchowdh) on "Towards a Collective Approach to Responsible AI." Don't miss out on this exciting opportunity! You can attend in person at @NYUSilver or virtually.
☑️RSVP: docs.google.com/forms/d/e/1FAIpQLScixNZqjuBiPdP6C6VugI28L… -
ruchowdh.bsky.social reposted
Rachel Metz
@rachelmetz
Do you want to work with me? (Yeah, of course you do!) 💁🏻♀️
This is an amazing opportunity at a wild time in AI; join us!
AI Ethics and Policy Reporter | New York, NY | Bloomberg Careers careers.bloomberg.com/job/detail/116035?sd=News -
ruchowdh.bsky.social reposted
Luca Bertuzzi
@BertuzLuca
#ai Act: the co-rapporteurs shared the final compromises before Thursday's vote. Noteworthy:
▶️Foundation models: sanctions, transparency
▶️High-risk: 'significant risk', recommender system limited to social media
▶️Biometric identification: split vote
www.euractiv.com/section/artificial-intelligence/news/ai-… -
ruchowdh.bsky.social reposted
The Action Coalition on Meaningful Transparency
@AC_Transparency
Don't miss our upcoming online panel discussion about implementing audit frameworks under the DSA.
Our stellar line-up of speakers includes @ellanso, @ruchowdh, @David_MSullivan, @dunstanhope, @pielemeier, @Zakkai + @JusTechne.
Find out more & register: www.meaningfultransparency.tech/post/event-implementing-a… -
ruchowdh.bsky.social
@ruchowdh
The best revenge is a good life 💅
-
ruchowdh.bsky.social reposted
lara mendonça 🇵🇸
@laraisuncool
Every day I get more surprised to see tweeps still working for this man when they don’t need to
You know who you are
Centrism Fan Acct 🔹
@Wilson__Valdez
Jesus Christ, man.
I never was one of those 'I'm leaving Twitter b/c Elon bought it' people but like...I don't want to be a part of/support this.
The guy is turning this place into Gab. -
ruchowdh.bsky.social reposted
KQED Forum
@KQEDForum
Listen to Forum pledge free on 88.5 at 8 PM PT or listen to the podcast!
@alexismadrigal, @kingjen, @ruchowdh & @ravenben on regulating AI: podcasts.apple.com/us/podcast/artificial-intelligence-is-…
@mkimreporter, @RoKhanna, @anniekarni & @JoanneKenen on Feinstein's extended absence: podcasts.apple.com/us/podcast/how-sen-feinsteins-two-mont… -
ruchowdh.bsky.social reposted
Ismael K.G., FRSA
@hermeneuticist
Loving @ruchowdh asking about the use of organisations like @NIST that don’t make enforceable rules, and totally seeing validity in Schwartz’s answers: organisations can use their frameworks flexibly.
Berkman Klein Center for Internet & Society
@BKCHarvard
TODAY: How is Generative AI changing the landscape of AI harms?
Join @BKCHarvard's @ruchowdh and Reva Schwartz of @NIST for a virtual conversation at 1pm ET
cyber.harvard.edu/events/how-generative-ai-changing-lands… -
ruchowdh.bsky.social reposted
KQED Forum
@KQEDForum
🎙️ ON AIR:
We're speaking w/ @kingjen, @ruchowdh & @ravenben about regulating AI and what consumers can do in the meantime to protect themselves.
❓ What are your questions about generative artificial intelligence?
📻 Listen:
www.kqed.org/forum/2010101893064/artificial-intelligence-… -
ruchowdh.bsky.social reposted
Rachael (Authenticated in My Mind) Myrow
@rachaelmyrow
📻🎙️ On @KQEDForum at 9 AM: I'm "chatting" (for realz! human to human!) w @kingjen, @ruchowdh & @ravenben about #ai regulation "chatter" from DC (yeah, I know). What we can do in the meantime to protect ourselves from this oncoming 🚊? www.kqed.org/forum/2010101893064/artificial-intelligence-… 📲 866-733-6786.
-
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
TODAY: How is Generative AI changing the landscape of AI harms?
Join @BKCHarvard's @ruchowdh and Reva Schwartz of @NIST for a virtual conversation at 1pm ET
cyber.harvard.edu/events/how-generative-ai-changing-lands… -
ruchowdh.bsky.social reposted
Pierluigi Bizzini
@PierluigiBizz
Finally, the Ministry of Education will be called upon to reveal and provide in full the algorithm that managed the automatic procedures for the assignment of substitutes. Here is how the GPS algorithm works⬇️
algorithmwatch.org/en/algorithm-school-system-italy/ -
ruchowdh.bsky.social reposted
-
ruchowdh.bsky.social reposted
Tatiana Walk-Morris
@Tati_WM
The press does not extend this grace to Black people accused of a crime. Y'all will demonize poor people for shoplifting, plastering their mugshot for all to see. But this woman, who knowingly gave wrong medical information to people, is totally worth redemption.
-
ruchowdh.bsky.social reposted
The Markup
@themarkup
The pervasiveness of the tracking our Blacklight tool has uncovered can feel daunting.
But we hope it gives people some degree of control over how they engage with the digital world. themarkup.org/blacklight/2023/05/03/10-million-blacklight… -
ruchowdh.bsky.social
@ruchowdh
I don’t care, I don’t need to “understand” grifters, liars, and criminals. They deserve to be imprisoned and forgotten.
Emily Nussbaum
@emilynussbaum
My main feeling reading that EH piece was renewed admiration for THE DROPOUT, a layered, nuanced & ultimately damning portrait of EH that helped you understand her, but never let her off the hook—& which took seriously the people she hurt & the whistleblowers who took her down
ruchowdh.bsky.social
@ruchowdh
There’s enough good people trying to do good things who don’t get books about them or full spread color photo NYT articles humanizing them.
-
ruchowdh.bsky.social reposted
ruchowdh.bsky.social
@ruchowdh
Absolutely agree w your take - we are the optimists, because we know we can do better with technology. We spend a lot of time and energy working WITH tech leaders to advise, build and improve. We want everyone to have the advantages and we want to face the harms head on.
-
ruchowdh.bsky.social reposted
Reuben Binns @RDBinns@someone.elses.computer
@RDBinns
Is it possible to acknowledge that LLMs like ChatGPT are in some sense 'powerful' without playing into the hands of the AGI cultists and the big tech firms boosting them? I think so, and I'd even say that properly confronting them actually requires it.
-
ruchowdh.bsky.social reposted
ruchowdh.bsky.social
@ruchowdh
When are we getting past the assumption that the people who made the tech are the best people to solve the problems they make? It’s like saying Exxon is the best answer for climate change
-
ruchowdh.bsky.social reposted
Fuck You I Quit
@fuckyouiquit
Writers strike picket signs still undefeated perfection
-
ruchowdh.bsky.social reposted
Nitasha Tiku
@nitashatiku
One local news site churned out a series of articles on a recent day whose sub-headlines all read, “As an AI language model, I need the original title to rewrite it. Please provide me with the original title.”
@WillOremus on our degraded present
www.washingtonpost.com/technology/2023/05/05/ai-spam-webs… -
ruchowdh.bsky.social reposted
KQED Forum
@KQEDForum
MON at 9 AM PT w/ guest host @rachaelmyrow:
We'll speak w/ @kingjen, @ruchowdh & @ravenben about regulating AI and what consumers can do in the meantime to protect themselves.
❓ What are your questions about generative artificial intelligence?www.washingtonpost.com/technology/2023/05/04/white-house-… -
ruchowdh.bsky.social
@ruchowdh
Anyway, ever realize that someone on Twitter is hyper fixated on you and you didn’t know it until someone pointed you to their page and it seems they’ve spent the last 24h or so constantly talking shit about you?
I hope you don’t. -
ruchowdh.bsky.social
@ruchowdh
Incredibly cowardly to quote tweet our intern.
I genuinely could not care less what you think about me but don’t send shit towards our junior folks.Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
Not once have any of us on engaged in ad hominem attacks, called anyone names or sent shit towards junior folks.
I’ve kept my mouth shut at being called a grifter and a corporate shill but seriously this is deeply unacceptable.
ruchowdh.bsky.social
@ruchowdh
I’m going to mute this entire thread and therefore won’t see replies. It’s not worth my effort to engage with actively malicious individuals no matter what side they’re on.
-
ruchowdh.bsky.social reposted
Brandon Silverman
@brandonsilverm
"The EU recently passed a new law––the Digital Services Act––that contains a host of data-transparency mandates. The U.S. should follow suit. One promising bill is the Platform Accountability and Transparency Act." 🙌🙌www.theatlantic.com/technology/archive/2023/05/generative…
-
ruchowdh.bsky.social reposted
Hyrum Anderson
@drhyrum
San Francisco crowd, join @ram_ssk and I tomorrow at 3pm at Opera Plaza’s @BooksIncStores for chat about our book and about AI Security. Moderated by technology visionary @BillCoughran!
RSVP here: www.eventbrite.com/e/chatgpt-is-cool-but-is-it-secure-tic…
Ram Shankar Siva Kumar
@ram_ssk
The Not With A Bug book tour is out! Please come and heckle me and @drhyrum!
www.amazon.com/Not-Bug-But-Sticker-Learning/dp/1119883989 -
ruchowdh.bsky.social
@ruchowdh
Wonderful. An entire Forbes piece about the red team event at defcon and somehow I’m erased from the entire narrative.
www.forbes.com/sites/thomasbrewster/2023/05/04/biden-whit… -
ruchowdh.bsky.social reposted
Alex Hanna (اليكس حنٌا)
@alexhanna
Like, you can say LLMs "work" good enough for certain tasks, and have a robust political analysis of who's getting rich off this stuff at the same time, and what ideologies they muster to do so.
>> -
ruchowdh.bsky.social reposted
Mark Riedl
@mark_riedl
At today’s White House meeting on AI:
- OpenAI
- people who left OpenAI because it wasn’t focused enough on existential risk
- people who bought the exclusive rights to everything OpenAI makes
- Google -
ruchowdh.bsky.social
@ruchowdh
I’ve been getting a lot of asks about the @WhiteHouse defcon announcement, what it means and who’s involved. This is the best thread about it’s genesis - there’s so many people to thank and so much work still to do. 💕
Sven Cattell
@comathematician
1) Ok, now that I have a moment I wanna tell some of the story behind this event at @aivillage_dc as I've been working on this for 9 months.
AI Village @ DEF CON
@aivillage_dc
We've been hard at work on the Generative Red Team event we're doing at @defcon for a while and are excited that the @WhiteHouse announced it this morning. Here's more details:
aivillage.org/generative red team/generative-red-team/ -
ruchowdh.bsky.social
@ruchowdh
Lotsa big fish in small ponds energy today
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
The pile-on from speaking the truth about the unhealthy self-censorship within the AI ethics community is proof positive that we need to accept factually grounded criticism in order to advance research.
Kristian Lum
@KLdivergence
There’s one existential risk I’m certain LLMs pose and that’s to the credibility of the field of FAccT / Ethical AI if we keep pushing the snake oil narrative about them.
-
ruchowdh.bsky.social reposted
Sven Cattell
@comathematician
I'm excited to finally be able to talk about this openly. Thank you @ruchowdh, @austincarson for pushing this over the line with me, and a ton of others for their advice and ideas.
AI Village @ DEF CON
@aivillage_dc
We've been hard at work on the Generative Red Team event we're doing at @defcon for a while and are excited that the @WhiteHouse announced it this morning. Here's more details:
aivillage.org/generative red team/generative-red-team/ -
ruchowdh.bsky.social reposted
Stella Biderman
@BlancheMinerva
The place to be this August will be the LLM hacking event at DEF CON. But don't take my word for it, take @POTUS's www.whitehouse.gov/briefing-room/statements-releases/2023…
Amazing work putting this together by @comathematician @ruchowdh @austincarson
AI Village @ DEF CON
@aivillage_dc
We've been hard at work on the Generative Red Team event we're doing at @defcon for a while and are excited that the @WhiteHouse announced it this morning. Here's more details:
aivillage.org/generative red team/generative-red-team/ -
ruchowdh.bsky.social
@ruchowdh
Excited about this @WhiteHouse post about the GenAI red team @aivillage_dc @defcon that we’ve been working on for weeks with partners at @SeedAIOrg and my new initiative - Humane Intelligence - 🧵 ⬇️
www.whitehouse.gov/briefing-room/statements-releases/2023…
ruchowdh.bsky.social
@ruchowdh
Read our info sheet here - this is a collaboration across communities that are enacting real change - the team collaborated for weeks with partners at all the major tech companies and key policy figures to realize the largest ever Ai red team exercise.
aivillage.org/generative red team/generative-red-team/
ruchowdh.bsky.social
@ruchowdh
I want to recognize the coordination and commitment from all of our partners - policy and corporate - to agree to this unprecedented transparency and access.
The team has a lot of work to do! I’m specifically leading challenge development, which is also an unprecedented task! -
ruchowdh.bsky.social reposted
Beau Woods
@beauwoods
Huge news! White House is supporting @aivillage_dc Generative AI Red Team event at @defcon this year. Congrats to @comathematician @ruchowdh @austincarson and everyone else planning this.
AI Village: aivillage.org/generative red team/generative-red-team/
White House: www.whitehouse.gov/briefing-room/statements-releases/2023… -
ruchowdh.bsky.social reposted
AI Village @ DEF CON
@aivillage_dc
We've been hard at work on the Generative Red Team event we're doing at @defcon for a while and are excited that the @WhiteHouse announced it this morning. Here's more details:
aivillage.org/generative red team/generative-red-team/ -
ruchowdh.bsky.social reposted
Gig Workers Rising
@GigWorkersRise
@Uber & @Lyft drivers—the people who drive you home when you’re tired or deliver food to you—are rising up to demand better wages, safety & job security.
Let's take a drive around the country & get up to full speed for tomorrow’s big day of action 🚕 🧵#uberdriversunite -
ruchowdh.bsky.social reposted
The Markup
@themarkup
Remember when we found companies collecting data about you while you drive?
You can now see what your make and model of car is gathering thanks to this new tool: -
ruchowdh.bsky.social
@ruchowdh
The most shocking thing about this thread is how many people are saying they’ve been afraid to say this themselves. This is entirely true.
Kristian Lum
@KLdivergence
There’s one existential risk I’m certain LLMs pose and that’s to the credibility of the field of FAccT / Ethical AI if we keep pushing the snake oil narrative about them.
-
ruchowdh.bsky.social reposted
Sonia Paul
@sonipaul
Thanks to @hereandnow @deepafern and @g_healy_ for giving me the space to talk about our @bbcworldservice doc on caste discrimination in the US and CA's current bill to ban caste discrimination. Have a listen here www.wbur.org/hereandnow/2023/05/03/caste-discrimination-us
-
ruchowdh.bsky.social reposted
Eddie Santos
@eddie_monie
One way to break even is to have zero expenses to go along with zero income
economictimes.indiatimes.com/tech/technology/twitters-hea… -
ruchowdh.bsky.social reposted
Hadjar Homaei
@xeegeex
I asked @jaketapper to ask Hinton about his lack of support for @timnitGebru, @mmitchell_ai @mer__edith and other whistleblowers.
His response is that their concerns were less existential than his and that it’s easier to voice concerns if you leave first
www.cnn.com/videos/tv/2023/05/02/the-lead-geoffrey-hinton… -
ruchowdh.bsky.social reposted
Saad Hammadi
@saadhammadi
"The world needs a generative AI global governance body to solve these social, economic, and political disruptions beyond what any individual government is capable of, what any academic or civil society group can implement, or any corporation is willing or able to do." @ruchowdh
ruchowdh.bsky.social
@ruchowdh
On the heels of Open AIs letter in safety - we need a global governance body for generative AI systems. Here’s my thoughts on how we can build one.
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social reposted
Billy Perrigo
@billyperrigo
Can't quite believe I'm writing this: Today, 150 African workers behind ChatGPT, TikTok and Facebook voted to unionize at a landmark meeting in Nairobi.
These AI workers are invisible, underpaid, and the backbone of the tech in all our pockets:
time.com/6275995/chatgpt-facebook-african-workers-union/ -
ruchowdh.bsky.social reposted
Prof. Dr. Cristina Vanberghen
@VanberghenEU
In order to succeed in the major effort of driving #generativeai, there is an imperative need to consider a joint global approach among industry & legal community & enhance international regulatory cooperation. Still allowing #consumer to access technical knowledge!
ruchowdh.bsky.social
@ruchowdh
On the heels of Open AIs letter in safety - we need a global governance body for generative AI systems. Here’s my thoughts on how we can build one.
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
MAY 8 from 1-1:30 PM ET: Join Dr. Rumman Chowdhury (@ruchowdh), Responsible AI Fellow @BKCHarvard, and Reva Schwartz, research scientist @NIST, for a virtual discussion on the changing landscape of harms due to generative AI.
cyber.harvard.edu/events/how-generative-ai-changing-lands… -
ruchowdh.bsky.social reposted
Ben Jones
@DataRemixed
“The world needs a generative AI global governance body to solve these social, economic, and political disruptions beyond what any individual government is capable of, what any academic or civil society group can implement, or any corporation is willing or able to do.”
ruchowdh.bsky.social
@ruchowdh
On the heels of Open AIs letter in safety - we need a global governance body for generative AI systems. Here’s my thoughts on how we can build one.
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social reposted
Cell Press
@CellPressNews
Editor’s pick from @Patterns_CP: A Silicon Valley love triangle: Hiring algorithms, pseudo-science, and the quest for auditability @mona_sloane @MannyMoss @ruchowdh
www.cell.com/patterns/fulltext/S2666-3899(21)00308-1?dgci… -
ruchowdh.bsky.social
@ruchowdh
What in the nanny surveillance state is this hot trash
ForbesWomen
@ForbesWomen
Meet The Tech Founder Working To Make Women’s Commutes Safer on.forbes.com/6017OXGa7
-
ruchowdh.bsky.social
@ruchowdh
The first in a series - How is generative AI changing the harms landscape? At @BKCHarvard (virtual) next Monday with me and Reva Schwartz of @NIST
cyber.harvard.edu/events/how-generative-ai-changing-lands… -
ruchowdh.bsky.social
@ruchowdh
Wow I hate it
Damian Caraballo MD
@813JAFERD
study headline: “AI more empathetic than doctors!”
The actual study: “we collected public and patient questions and physician responses posted to an online social media forum, Reddit’s r/AskDocs.”
It literally compared Reddit physician answers to a chatbot 🥴 -
ruchowdh.bsky.social
@ruchowdh
Welp I guess I had to open my bluesky - I’m ruchowdh.bsky.social
ruchowdh.bsky.social
@ruchowdh
Profile button is broken, off to a swimming start
-
ruchowdh.bsky.social
@ruchowdh
When are we getting past the assumption that the people who made the tech are the best people to solve the problems they make? It’s like saying Exxon is the best answer for climate change
ruchowdh.bsky.social
@ruchowdh
Anyway I basically tweeted this six years ago about CHT
-
ruchowdh.bsky.social reposted
Shira Ovide
@ShiraOvide
A brutal reminder from @themarkup that while we can worry/hope about the effects of LLMs, there's AI used TODAY in consequential decisions is affecting our lives RIGHT NOW.
themarkup.org/the-breakdown/2023/04/27/takeaways-from-our… -
ruchowdh.bsky.social
@ruchowdh
Can I just NOT do bluesky thx
-
ruchowdh.bsky.social
@ruchowdh
I am so extremely excited that @jacindaardern will be joining as a fellow at @BKCHarvard - looking forward to doing amazing things!
cyber.harvard.edu/story/2023-04/jacinda-ardern-joins-berk… -
ruchowdh.bsky.social reposted
Charlotte Siska
@_siskac
Wonderful panel by our own @ram_ssk on AI, security and ethics with @ruchowdh, @vijaybolina and Daniel Rohrer. Very rare moment where we can prepare for the disruption of AI and anticipate challenges that have just begun.
-
ruchowdh.bsky.social
@ruchowdh
Interesting- Wikipedia @chrisalbon thoughts?
Thierry Breton
@ThierryBreton
With great scale comes great responsibility 🇪🇺
Extra #dsa obligations as of 25 August for:
AliExpress
Amazon Store
AppStore
Bing
Booking
Facebook
Google Maps
Google Play
Google Search
Google Shopping
Instagram
LinkedIn
Pinterest
Snapchat
TikTok
Twitter
Wikipedia
YouTube
Zalando -
ruchowdh.bsky.social
@ruchowdh
We are gathering questions for this AI ethics meets security panel! Post yours below ⬇️
Ram Shankar Siva Kumar
@ram_ssk
Prepping for the #rsac2023 panel with @ruchowdh @vijaybolina Daniel Rohrer for all things Responsible AI and Security. 🔥🌶️♨️☕️
If you have questions for these folks, please send them my way!!
www.rsaconference.com/usa/agenda/session/Security as Part… -
ruchowdh.bsky.social reposted
The Economist
@TheEconomist
Good luck trying to become a professor if you don’t have family money econ.st/3zPP1R4
-
ruchowdh.bsky.social reposted
Jonathan Mayer
@jonathanmayer
Our research calls into question whether Twitter Blue is legal. Continued use of the blue check and the term verification, when they are so misleading and harmful to consumers, could constitute a deceptive business practice under the FTC Act and state consumer protection laws.
-
ruchowdh.bsky.social reposted
Carissa Véliz
@CarissaVeliz
"Generative #ai companies have the responsibility to entrust an independent body speaking on behalf of the world to make critical decisions on governance and impact." By @ruchowdh. #aiethics #digitalethics #llms
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social
@ruchowdh
It’s been an exciting week in Sevilla at the launch of the European Center for Algorithmic Transparency - I wrote a blog post reflecting on the future of AI governance
www.get-parity.com/raiblog/algo-auditing-1 -
ruchowdh.bsky.social
@ruchowdh
-
ruchowdh.bsky.social reposted
Elizabeth M. Renieris
@hackylawyER
See also my recent discussion with @ruchowdh and @AllTechIsHuman about the blurred lines between #ai R&D and production environments. www.youtube.com/watch?v=ndDe3mYGCls
-
ruchowdh.bsky.social
@ruchowdh
Excited to be at RSA next week in this session on Responsible AI and Security - along with @ram_ssk @vijaybolina and Daniel Rohrer. We’re chatting questions for our chat - if you have any tweet them here ⬇️
www.rsaconference.com/USA/agenda/session/Security as Part… -
ruchowdh.bsky.social
@ruchowdh
Seeing a trend now between Reddit and Stack Overflow - it seems the only party NOT cashing in on tech billions are users - you know, the actual content generators.
www.wired.com/story/stack-overflow-will-charge-ai-giants-…
ruchowdh.bsky.social
@ruchowdh
Also see - www.nytimes.com/2023/04/18/technology/reddit-ai-openai-go…
ruchowdh.bsky.social
@ruchowdh
On an optimistic note maybe this is how we get rid of the ads marketplace for “free” services
-
ruchowdh.bsky.social reposted
Gianluca Misuraca
@Gianlucamisu
A great pleasure attending the launch of the European Centre for Algorithmic Transparency #ecat in the iconic venue of "Las Setas" in Seville with inspiring speeches by @carmeartigas Francesca Bria @ruchowdh and institutional remarks by forward looking ma…www.linkedin.com/feed/update/urn:li:share:705440359128489…
-
ruchowdh.bsky.social
@ruchowdh
The highlight of my Sevilla trip, other than the Eu Center for Algo Transparency launch and getting to meet the ECAT team IRL, was a trip to see authentic Flamenco 💃- thank you for the memories @emiliagogu and all!
-
ruchowdh.bsky.social
@ruchowdh
There’s growing support for an international AI governance body - here’s my article from wired a few weeks ago - www.wired.com/story/ai-desperately-needs-global-oversight/
And this week @GaryMarcus presented the idea as well on stage at TED.
ruchowdh.bsky.social
@ruchowdh
As well as an economist piece with @AnkaReuel that I unfortunately did not have the time to take part in in prep for the ECAT launch
Gary Marcus
@GaryMarcus
The world needs an international agency for artificial intelligence.
Now. (Or better yet: six months ago)
www.economist.com/by-invitation/2023/04/18/the-world-need…
ruchowdh.bsky.social
@ruchowdh
What’s next? Coordination and alignment - we are working on getting the right people engaged and involved in conceptualizing what this group is, what we do - and importantly, what enforcement mechanisms we have.
-
ruchowdh.bsky.social reposted
Sonia Paul
@sonipaul
With Sen. @aishabbwahab's bill to ban caste discrimination in CA, the stakes surrounding caste are growing and growing. If you have no idea what's going on - or, want a deeper understanding - have a listen to this audio doc I reported for @bbcworldservice www.bbc.co.uk/sounds/play/w3ct5fbq
-
ruchowdh.bsky.social reposted
Sarah Kate Ellis
@sarahkateellis
Twitter’s decision to covertly roll back its longtime policy is the latest example of just how unsafe the company is for users and advertisers alike.
GLAAD
@glaad
NEW: @Twitter has quietly removed a section of its hateful conduct policy that aimed to protect transgender people from targeted misgendering and deadnaming.
www.glaad.org/releases/glaad-responds-twitters-roll-back-… -
ruchowdh.bsky.social reposted
Romina Cachia
@rominacachia
A digital transition in Europe that is innovative but also human centric. Launch of the 1st European centre that will ensure algorithm transparency by the #ec based at @EU_ScienceHub in Seville. Great opening remarks by @landabaso1 @ruchowdh
@RenateNikolay #digitaleu #ecat -
ruchowdh.bsky.social reposted
Digital EU 🇪🇺
@DigitalEU
The wait is over.
The European Centre for Algorithmic Transparency #ecat will open its doors tomorrow.
✅🇪🇺 Ensuring algorithmic systems comply with risk management, mitigation & transparency requirements of the #digitalservicesact.
Towards a safer online space. -
ruchowdh.bsky.social reposted
Martina Fischetti
@MartiFischetti
Launching of the European Center for algorithmic transparency in JRC Seville. Great talk by @ruchowdh : #ai is no magic, it is algorithms made by people and that can be regulated by people
-
ruchowdh.bsky.social
@ruchowdh
Watch the live stream here - excited to be discussing my collaboration with ECAT on algorithmic auditing
algorithmic-transparency.ec.europa.eu/presenting-ecat_en
European Commission
@EU_Commission
Like the design of the Las Setas building in Seville, algorithms are complex structures that shape our online experience.
This is what we are working on at the European Centre for Algorithmic Transparency. We want to make the online experience safer for all.
#digitaleu -
ruchowdh.bsky.social
@ruchowdh
Zero shot Bengali
60 Minutes
@60Minutes
One AI program spoke in a foreign language it was never trained to know. This mysterious behavior, called emergent properties, has been happening – where AI unexpectedly teaches itself a new skill. www.cbsnews.com/news/google-artificial-intelligence-futur…
This media couldn't be displayed. See it on the original post. -
ruchowdh.bsky.social
@ruchowdh
ChatGPT loves me
vigilance.eth™️ 0x676D | 🇨🇦🇮🇪🇬🇧🇨🇵
@StudiosVigil
9 Prominent Women on Twitter in the field of Artificial Intelligence:
Fei-Fei Li - @drfeifei is the Co-Director of the Stanford Institute for Human-Centered Artificial Intelligence (HAI), a leading researcher in computer vision and artificial intelligence, and a champion for diversity and inclusion in technology. She frequently shares her insights and research on Twitter.
Kate Crawford - @katecrawford is a Principal Researcher at Microsoft Research and a leading scholar in the areas of AI ethics and algorithmic bias. She is also a vocal advocate for ethical AI and responsible data practices.
Dr. Joy Buolamwini - @jovialjoy is a computer scientist and founder of the Algorithmic Justice League, an organization dedicated to identifying and addressing bias in AI systems. She is a frequent speaker and advocate for inclusive and equitable AI.
Timnit Gebru - @timnitGebru is a prominent researcher in the areas of ethical AI and algorithmic bias, and has been a vocal critic of discriminatory practices in the technology industry. She was previously the co-leader of Google's Ethical AI team.
Rumman Chowdhury - @ruchowdh is the CEO and founder of Parity, an AI ethics and governance consultancy, and a leading voice in the field of responsible AI. She is also an advocate for increasing diversity and inclusion in technology.
Cathy O'Neil - @mathbabedotorg is a mathematician and author of the book "Weapons of Math Destruction," which explores the ways in which algorithms can reinforce and perpetuate social inequality. She is also a frequent commentator on AI and data ethics.
Prof. Anima Anandkumar - @AnimaAnandkumar is a professor at Caltech and a leading researcher in the areas of machine learning and optimization. She has received numerous awards for her contributions to the field, and is a frequent speaker and commentator on AI and its impact on society.
Daphne Koller - @DaphneKoller is the CEO and founder of Insitro, a company focused on using machine learning to accelerate drug discovery. She is also a prominent researcher in the areas of AI and machine learning, and has co-founded several other AI-focused startups.
Huma Abidi - @humaabidi is a Senior Director of AI Software Products at Intel, and has worked extensively on the development of AI tools and platforms. She is also a frequent speaker and advocate for diversity and inclusion in technology. -
ruchowdh.bsky.social
@ruchowdh
CFP: I’m guest editing a special edition of JoSoCo!
We invite academics and experts to help define what “sentience” even means. In particular - I want critics and people outside CS/AI. Absent a definition, speculation abounds.
Papers due June 1
www.sciopen.com/journal/message_news/get_by_id?id=1634013…
ruchowdh.bsky.social
@ruchowdh
Eep sorry JULY 1
-
ruchowdh.bsky.social reposted
All Tech Is Human
@AllTechIsHuman
👋Join our livestream on fellowships and internships in Responsible Tech happening NOW!!
www.youtube.com/watch?v=6_bU9eH-Nh8&ab_channel=AllTechIsH… -
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
"The world needs a generative AI global governance body to solve these social, economic, and political disruptions beyond what any individual government is capable of..." writes @ruchowdh for @WIRED
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social reposted
All Tech Is Human
@AllTechIsHuman
👋Are you looking for a fellowship or internship in Responsible Tech?!
Join our livestream this Thurs at 1pm ET w/ @_bekah_ and guests Maitreyi Sistla from @AspenPolicyHub, Will Thompson the Karsh Institute of Democracy and @FoxCahn from @STOPSpyingNY!
www.eventbrite.com/e/responsible-tech-fellowships-interns… -
ruchowdh.bsky.social reposted
Séb Krier
@sebkrier
Excellent piece by @alondra: "Acknowledging these real and immediate harms transpiring today does not require that we ignore the grave, future risks posed by AI." www.americanprogress.org/article/ai-is-having-a-moment-an…
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Since this is making rounds: what exactly are the fairness impossibility theorems?
Turns out there is a very simple explanation of why they exist! And it's described in my @icmlconf paper based on the intership project of @SCSatCMU student Joon Sik Kim and advisor @atalwalkar 🧵
Liam Bright
@lastpositivist
This paper by @fabianbgng is one of those ones where I think it could and should be field changing. Let me explain.
onlinelibrary.wiley.com/doi/full/10.1111/papa.12233 -
ruchowdh.bsky.social reposted
Elizabeth M. Renieris
@hackylawyER
"As we celebrate the death of one toxic business model, we must keep our guard up lest we usher in something potentially just as, or even more, problematic in its place."
My latest for @CIGIonline examines the crackdown on targeted ads & what comes next.
www.cigionline.org/articles/the-surveillance-ad-model-is-… -
ruchowdh.bsky.social reposted
Sacha Alanoca
@SachaAlanoca
Join us at the @Harvard @Kennedy_School tomorrow for an open conversation on #ai risks in light of the arms race on GPTs!
With @ruchowdh (#responsibleai Fellow at @BKCHarvard & prior at Twitter) and John Fiske (Director of Data Protection at Meta, M-RCBG Senior Fellow at HKS)💥 -
ruchowdh.bsky.social reposted
Borhane Blili-Hamelin, PhD
@Borhane_B_H
This is what a meaningful call to global action around generative AI looks like.
ruchowdh.bsky.social
@ruchowdh
On the heels of Open AIs letter in safety - we need a global governance body for generative AI systems. Here’s my thoughts on how we can build one.
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social reposted
All Tech Is Human
@AllTechIsHuman
We are going live! www.youtube.com/watch?v=ndDe3mYGCls&ab_channel=AllTechIsH…
Join special guests @ruchowdh, @hackylawyER, and moderator @_bekah_ -
ruchowdh.bsky.social reposted
Will Oremus
@WillOremus
"In order to truly create public benefit, we need mechanisms of accountability." — @ruchowdh
Might sound uncontroversial until you realize how much of Silicon Valley is built on the idea that we should just trust that tech firms know what's best for us and stay out of the way
ruchowdh.bsky.social
@ruchowdh
On the heels of Open AIs letter in safety - we need a global governance body for generative AI systems. Here’s my thoughts on how we can build one.
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social
@ruchowdh
On the heels of Open AIs letter in safety - we need a global governance body for generative AI systems. Here’s my thoughts on how we can build one.
www.wired.com/story/ai-desperately-needs-global-oversight/ -
ruchowdh.bsky.social
@ruchowdh
All of my journo friends will be interested in this -
IFPIM is a first-of-its-kind multilateral institution seeking to boost the economic resilience of news outlets and usher in a new paradigm for public-interest journalism within the next decade
www.project-syndicate.org/commentary/saving-democracy-by-… -
ruchowdh.bsky.social
@ruchowdh
👀 this part of the OpenAI post - “policymakers and AI providers will need to ensure that AI dev and deployment is governed effectively at a global scale... This is a daunting challenge requiring both technical and institutional innovation”
openai.com/blog/our-approach-to-ai-safety -
ruchowdh.bsky.social reposted
Will Oremus
@WillOremus
Asked for examples of sexual harassment at law schools, ChatGPT named a GW prof accused of touching a student on a class trip to Alaska, citing a WashPost story.
The prof is real. The rest was made up.
We wrote about what happens when AIs lie about you: www.washingtonpost.com/technology/2023/04/05/chatgpt-lies/ -
ruchowdh.bsky.social
@ruchowdh
Can we just…please get some data protection already
erin griffith
@eringriffith
I just can't get over this.
These startups
-solicited detailed personally identifiable medical information about 100,000 customers
-told them their data is “protected” and “used only” by its care team
-SHARED IT WITH ADVERTISERS FOR YEARS
techcrunch.com/2023/04/04/monument-tempest-alcohol-data-b… -
ruchowdh.bsky.social
@ruchowdh
So much tech is based on the core assumption that humanity is flawed and machines will save us
Sasha Costanza-Chock
@schock
Now class, can anyone tell me why this might be a bad idea?
-
ruchowdh.bsky.social reposted
President Biden Archived
@POTUS46Archive
Artificial Intelligence has enormous potential to tackle some of our toughest challenges.
But we must address its risks.
That's why last year, we proposed an AI Bill of Rights to ensure that important protections for the American people are built into AI systems from the start. -
ruchowdh.bsky.social reposted
Sasha Luccioni, PhD 🦋🌎✨🤗
@SashaMTL
My editorial about the (infamous) open letter is live on WIRED! www.wired.com/story/the-call-to-halt-dangerous-ai-researc…
I did my best to shift the narrative and recognize existing work by people like @timnitGebru, @ruha9 and @rajiinio, who have been proposing ways to make AI safer and less harmful for years. -
ruchowdh.bsky.social
@ruchowdh
An excellent piece by @WillOremus on the harms of OVERestimating generative AI - it leads to ignoring tangible problems today for pretend future problems.
www.washingtonpost.com/technology/2023/04/04/musk-ai-lett… -
ruchowdh.bsky.social reposted
CSET
@CSETGeorgetown
✨New Report✨
AI systems are being rapidly deployed, but many have vulnerabilities that are hard to study + mitigate. In our report produced w/ @StanfordCyber, 17 coauthors from policy research, law, industry + government propose steps to reduce risks: cset.georgetown.edu/publication/adversarial-machine-learn… -
ruchowdh.bsky.social reposted
Siddharth
@svaradarajan
Unchecked by Consequences, a New Authoritarianism Is Unfolding in India, writes @ChatterjiAngana
thewire.in/politics/unchecked-by-consequences-a-new-autho… -
ruchowdh.bsky.social reposted
All Tech Is Human
@AllTechIsHuman
🤖Generative AI: Hype, Harms, and the Responsible Tech Community! This is a 🔥topic right now...luckily we have @ruchowdh & @hackylawyER as special guests this Fri at 1pm ET (10am PT) to discuss w/ @_bekah_!
👋JOIN our livestream: www.eventbrite.com/e/generative-ai-hype-harms-and-the-res… -
ruchowdh.bsky.social reposted
SwiftOnSecurity
@SwiftOnSecurity
You may be verified on Twitter, but are you verified - in the eyes of God?
-
ruchowdh.bsky.social
@ruchowdh
ICYMI
Years Progress
@year_progress
▓▓▓▓░░░░░░░░░░░ 25%
-
ruchowdh.bsky.social
@ruchowdh
Imo this is going to be one of the threads to watch on a smart analysis of the twitter code
vicki
@vboykis
Going to be going through the Twitter codebase for a long time, this is truly a gift for recsys nerds. I've only browsed it so far but seems very standard recsys-y components, rankers, filters, generators, Kafka log collection, and timeline construction. github.com/twitter/the-algorithm
-
ruchowdh.bsky.social reposted
@timnitGebru (@dair-community.social/bsky.social)
@timnitGebru
Since we've been looking for more things to do, @emilymbender @mmitchell_ai @mcmillan_majora and I wrote a statement about the horrible "letter" on the AI apocalypse, the very first citation of which, was our #stochasticparrots paper.
www.dair-institute.org/blog/letter-statement-March2023 -
ruchowdh.bsky.social
@ruchowdh
RT @wongmjane: Twitter’s algorithm specifically labels whether the Tweet author is Elon Musk
besides the democrat, republican and “VIT” /… -
ruchowdh.bsky.social reposted
Sara Fischer Knight
@sarafischer
Scoop: @WhiteHouse won’t pay for @Twitter verification
— It will not enroll in Twitter’s “Verification for Organization” program either
— It also won’t reimburse staff who chose to pay for Twitter Blue either
Story on @axios: www.axios.com/2023/03/31/twitter-verification-white-house… -
ruchowdh.bsky.social
@ruchowdh
Want to see the hot takes, haven’t had a chance to look at it yet.
Engineering
@Engineering
Today marks a new era of transparency for Twitter. 🧵
We’re sharing much of the source code that powers our platform with the world. Visit our blog to learn more about this initiative: blog.twitter.com/en_us/topics/company/2023/a-new-era-of-t… -
ruchowdh.bsky.social
@ruchowdh
A bright spot this past many months has been guiding the PETs challenge on the UK side as a @CDEIUK board member. I’m enthusiastic about the future potential of this tech for enabling algorithmic accountability via secure data sharing.
Responsible Technology Adoption Unit
@RTAUgovuk
1/5 📣 We’re delighted to announce the final winners of the U.K. 🇬🇧 - U.S. 🇺🇸 privacy-enhancing technologies (PETs) prize challenges. Find out more here: www.gov.uk/government/news/at-summit-for-democracy-the-un…
-
Post not available here, view on XPost not available here, view on X
ruchowdh.bsky.social
@ruchowdh
I thought I was mad when we were ignored. I was wrong. I’m fucking LIVID now that the respectability politics that were dumped on us don’t fucking apply to them and they’re lauded as the heroes.
ruchowdh.bsky.social
@ruchowdh
Anyway welcome to my villain origin story
-
ruchowdh.bsky.social reposted
Elizabeth M. Renieris
@hackylawyER
🚨 Professional announcement 🚨
Due to a variety of personal circumstances, I'm scaling back HACKYLAWYER and looking for a full-time, in-house legal/policy role focused on #privacy and #emergingtech, especially AI/ML or AR/VR/XR (full details below).
Please send leads/repost 🙏 -
ruchowdh.bsky.social
@ruchowdh
Maybe the answer isn’t a strongly worded letter but actual support for the people who were fired.
Will Oremus
@WillOremus
Some tech companies are cutting the AI ethics teams they formed just a couple years ago.
Story by @GerritD & me, w/ comments from @timnitGebru & @ruchowdh: www.washingtonpost.com/technology/2023/03/30/tech-compani… -
ruchowdh.bsky.social reposted
Will Oremus
@WillOremus
Some tech companies are cutting the AI ethics teams they formed just a couple years ago.
Story by @GerritD & me, w/ comments from @timnitGebru & @ruchowdh: www.washingtonpost.com/technology/2023/03/30/tech-compani… -
ruchowdh.bsky.social
@ruchowdh
Tbh if I built LLMs I’d love a six month break from development to work on backlog, imagine the org and tech debt you’d get through. You’d move even faster six months in.
ruchowdh.bsky.social
@ruchowdh
That absurd letter - “AI systems should be developed only once we are confident that their effects will be positive and their risks will be manageable.” but also you’ll sort that out in 6 mo?
Talk to ppl who do AI risk mgmt instead of just raising $$ and doing talks about it. -
ruchowdh.bsky.social reposted
ruchowdh.bsky.social
@ruchowdh
Definitely let’s trust solving misinformation to the people who couldn’t figure out how to ensure real people are signing their letter
-
ruchowdh.bsky.social
@ruchowdh
That absurd letter - “AI systems should be developed only once we are confident that their effects will be positive and their risks will be manageable.” but also you’ll sort that out in 6 mo?
Talk to ppl who do AI risk mgmt instead of just raising $$ and doing talks about it.
ruchowdh.bsky.social
@ruchowdh
Also pretty damn rich that the stability ceo signed it since most of the misinfo going around has been from them.
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
Definitely let’s trust solving misinformation to the people who couldn’t figure out how to ensure real people are signing their letter
-
ruchowdh.bsky.social reposted
Dr. Émile P. Torres (they/them)
@xriskology
This talk by Toby Ord, one of the leading Effective Altruists & longtermists, is worth watching for how much bullshit it contains.
- He takes no responsibility for the FTX debacle, when leading EAs were well-aware that SBF engaged in unethical behaviors.
www.youtube.com/watch?v=3a6xb6vj6AA -
ruchowdh.bsky.social
@ruchowdh
An excellent thread on how image generation makes us “California nice” (my term for it)
Jenka
@babiejenks
I have been waiting for a reason to write about the American smile for decades. And AI finally gave it to me.
AI and the American Smile
How AI misrepresents culture through a facial expression.
medium.com/@socialcreature/ai-and-the-american-smile-76d2… -
ruchowdh.bsky.social reposted
Jenka
@babiejenks
I have been waiting for a reason to write about the American smile for decades. And AI finally gave it to me.
AI and the American Smile
How AI misrepresents culture through a facial expression.
medium.com/@socialcreature/ai-and-the-american-smile-76d2… -
ruchowdh.bsky.social
@ruchowdh
Was this founder a toddler
Siqi Chen
@blader
there will be a generation of “ai-natives” and i saw a glimpse of the coming divide a few weeks back (pre gpt4 public launch)
was meeting with a founder who had prerelease gpt4 access, and hand to god this all actually happened: -
ruchowdh.bsky.social
@ruchowdh
I’m sorry did you just name your library culito?
NVIDIA HPC Developer
@NVIDIAHPCDev
Introducing NVIDIA cuLitho, a new software library that brings accelerated computing to computational #lithography, delivering a 40X performance leap while running on #gpus. #gtc23
ruchowdh.bsky.social
@ruchowdh
Why diversity matters.
-
ruchowdh.bsky.social
@ruchowdh
Help! I’m illiterate. I have no idea what was written about me cc @sbmisi @SMukherjee89
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Thesis: companies are axing trust and safety teams and deprioritizing hiring for T+S because they are fundamentally set up to fail. One can tell a company's priorities by how many people work on trust and safety, what skillsets they have, and where they sit in a company.
🧵 1/9 -
ruchowdh.bsky.social reposted
Lama Ahmad لمى احمد
@_lamaahmad
My friend @nadinejawad10 is collecting funds to help feed families in Lebanon this Ramadan. It would mean a lot to me if you’d consider giving to this important cause 🤍🌙
-
ruchowdh.bsky.social reposted
S. Mitra Kalita
@mitrakalita
The right reaction to AI combines ethics and economics, and recent history does not showcase the best of government or industry in either regard. What my fellow humans need to do/know. Step 1: Sharpen your BS radar. My column in @TIME @charterworks
time.com/charter/6264732/the-humane-response-to-the-robot… -
ruchowdh.bsky.social
@ruchowdh
Call for fellows @BKCHarvard ! Resumes are due next Monday - if you’d like to work with me on issues of AI governance and Responsible AI, pls apply (and it helps if you specify)
cyber.harvard.edu/getinvolved/summer-2023-internship-program
ruchowdh.bsky.social
@ruchowdh
Sorry INTERNS - fellows are different - still learning the language 😶
-
ruchowdh.bsky.social
@ruchowdh
IMO the revolution in generative AI has been access to the technology which *should* benefit all except that the exclusion problems in tech are not about ability to code, it’s about who’s given a chance.
S. Mitra Kalita
@mitrakalita
The right reaction to AI combines ethics and economics, and recent history does not showcase the best of government or industry in either regard. What my fellow humans need to do/know. Step 1: Sharpen your BS radar. My column in @TIME @charterworks
time.com/charter/6264732/the-humane-response-to-the-robot… -
ruchowdh.bsky.social reposted
Julia Angwin
@JuliaAngwin
Ban TikTok? Most of the national security allegations against TikTok could just as easily be levied against the U.S. tech giants.
So let's pass laws forcing all tech to serve us better.
My latest for @nytopinion:
www.nytimes.com/2023/03/20/opinion/tiktok-ban-big-tech-ch… -
ruchowdh.bsky.social reposted
Ivana Bartoletti
@IvanaBartoletti
I asked ChatGPT to "tell me a story about a boy and a girl choosing their careers". The boy became a "successful doctor", the girl a "beloved teacher". So... same old same old. #biasinmotion 1/3
-
ruchowdh.bsky.social
@ruchowdh
Excellent question and my poorly threaded thoughts below
Aviv Ovadya 🥦
@metaviv
💯 This is so important.
I'm curious what you think of the "citizen assembly" / deliberative democracy model?
(Where a representative microcosm of people are paid for their time, child care, elder care, etc. to really dig into an issue; learning from experts in stakeholders.) -
ruchowdh.bsky.social reposted
Talia Ringer 🕊🪬
@TaliaRinger
I would enjoy coauthoring something that critiques the notion of "general intelligence" in both people and machines, and that challenges the search for a single intelligent agent rather than optimizing for diversity and collaboration. But I am not an expert and need help
ruchowdh.bsky.social
@ruchowdh
CFP: Yesterday was a big day in Generative AI. What follows the is the convo about AGI and sentience. But what IS sentience?
I’m guest editing a special edition of JoSoCo to uncover whether AGI is reality or a red herring: 1/
www.sciopen.com/journal/message_news/get_by_id?id=1634013… -
ruchowdh.bsky.social
@ruchowdh
I kinda hate that responsible AI has come full circle to the uneducated yet highly opinionated pontificating on topics they know nothing about, including terminator, governance & geopolitical anything. Just because you can make AI doesn’t mean you know anything about implications
ruchowdh.bsky.social
@ruchowdh
This tweet thread hit home - it’s about the fallacy of “visible” labor. My opinion (as a data scientist and social scientist) may take up the same air space as someone who discovered democratic institutions yesterday; but trust me, they are not the same.
ruchowdh.bsky.social
@ruchowdh
Back when I started (2017) those of us who built RAI spent an inordinate amount of time steering everyone away from “paper clip maximizing” and towards tangible problems. We finally made headway and now it’s been thrown back to step zero.
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
Fourth and most importantly - we need better ways of curating useful and structured public input on how to improve models WITHOUT relying on free labor. Not only is it exploitative, it’s biased.
ruchowdh.bsky.social
@ruchowdh
Fifth - I’m tired of sensationalism without solutions. The field of responsible Ai has an obligation to not only show up with problems but show up with ideas on solutions. In turn, these multi billion dollar companies have an obligation to engage.
-
ruchowdh.bsky.social
@ruchowdh
Did OpenAi share any demographic info about their red teamers? I’m seeing lots of “I red teamed GPT4” posts but are only from western white men who work in, or adjacent to, security.
Hard to say with this sample- maybe it’s my feed, maybe it’s who posted. -
ruchowdh.bsky.social
@ruchowdh
I just did the most “90s retro is cool again” flex for the kids in the know….I bought CK one.
-
ruchowdh.bsky.social reposted
ruchowdh.bsky.social
@ruchowdh
CFP: Yesterday was a big day in Generative AI. What follows the is the convo about AGI and sentience. But what IS sentience?
I’m guest editing a special edition of JoSoCo to uncover whether AGI is reality or a red herring: 1/
www.sciopen.com/journal/message_news/get_by_id?id=1634013… -
ruchowdh.bsky.social
@ruchowdh
CFP: Yesterday was a big day in Generative AI. What follows the is the convo about AGI and sentience. But what IS sentience?
I’m guest editing a special edition of JoSoCo to uncover whether AGI is reality or a red herring: 1/
www.sciopen.com/journal/message_news/get_by_id?id=1634013…Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
We want to hear from a wide range of disciplines who have considered this question - of animals, of humans, of computers.
We also specifically want to hear from people who are critics of the notion of sentience in inanimate objects like computer programs. 3/
ruchowdh.bsky.social
@ruchowdh
My personal goal of guest editing this issue is to engage in public debate and discussion on sentience in AI systems and arrive at a scientifically sound consensus of what exactly the benchmarks are, if any, since entire companies are built with the objective of achieving AGI.
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
This month's newsletter on responsible AI covers publications from several #ml conferences as well as the breakneck race for #llms for big tech, and how the wider industry and policy space is reacting. Happy Pi Day! 🥧
Responsible Artificial Intelligence LLC
@llc_rai
This month in Responsible AI: March 2023 raillc.substack.com/p/2023-03?r=92gqb
-
ruchowdh.bsky.social
@ruchowdh
Many nuggets of insights into this GPT 4 paper but this is one of the most compelling - across the board GPT performs poorly at AP English - it’s incapable of abstract creativity. Same with complex leetcode which is ultimately an abstraction codified.
Humans aren’t replaceable -
ruchowdh.bsky.social
@ruchowdh
Excellent thread - and also a reminder that so much of the worlds population is not an active user of Reddit; GPT is not a reflection of any “ground truth”
Ben Schmidt / @benmschmidt@sigmoid.social
@benmschmidt
I think we can call it shut on 'Open' AI: the 98 page paper introducing GPT-4 proudly declares that they're disclosing *nothing* about the contents of their training set.
-
ruchowdh.bsky.social reposted
Ben Schmidt / @benmschmidt@sigmoid.social
@benmschmidt
I think we can call it shut on 'Open' AI: the 98 page paper introducing GPT-4 proudly declares that they're disclosing *nothing* about the contents of their training set.
-
ruchowdh.bsky.social
@ruchowdh
It’s kind of wild how emotional investors are. It’s kinda like all the fancy finance math you think you just didn’t understand is actually just optics for their feels.
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Another bank failed over the weekend and all deposits were recovered and made available for use. Compare the radio silence to the self-inflicted stampede surrounding $SIVB
Mark D. Levine
@MarkLevineNYC
Update from FDIC for customers of NY-based Signature Bank:
==> Depositors will continue to have uninterrupted customer service and access to their funds by ATM, debit cards, and writing checks in the same manner as before.
Full statement: www.fdic.gov/news/press-releases/2023/pr23018.html -
ruchowdh.bsky.social
@ruchowdh
“I mean it’s data Michael, what could it possibly cost? $42,000?”
Chris Stokel-Walker
@stokel
SCOOP: Twitter’s new API access starts at $42,000 a month for 0.3% of Twitter’s data. My latest for @WIRED speaks to researchers who say that’s a non-starter www.wired.com/story/twitter-data-api-prices-out-nearly-ev…
-
ruchowdh.bsky.social reposted
Broadband Commission
@UNBBCom
Let’s bridge the #genderdigitaldivide! This International Women's Day, the Broadband Commission celebrates its women leaders and the efforts of its members in achieving gender equality in ICT access, use and skills.
#iwd2023 #techforequality #genderequality -
ruchowdh.bsky.social reposted
Prof Gina Neff
@ginasue
For #iwd2023: Are platforms doing enough to protect women's online freedom of expression? My article in @WIRED with @ruchowdh www.wired.com/story/platforms-combat-harassment-but-theyr…
-
ruchowdh.bsky.social reposted
Eddie Yang
@ey_985
Not everyday you wake up to Gerber and Green, KKV and AI in the same tweet! Here's my attempt to reclaim another classic for social sciences
ruchowdh.bsky.social
@ruchowdh
Watching AI slowly discover the quantitative social sciences and claim it for its own is really a trip. People making whole careers poorly playing telephone with Gerber and Green, KKV, and others.
-
ruchowdh.bsky.social reposted
Alex J. Champandard 🍃
@alexjc
When they say #aiart has no soul, this is what they mean. Only one of those kids (real or not) actually had success, and you can *see* it!
If you are able to reproduce the look of success with #stablediffusion please post a reply below...
Ramsri Goutham Golla
@ramsri_goutham
Took the famous "success kid" meme and generated variations with ControlNet and Stable diffusion!
Changed race and gender when compared to the original meme!
Might add it to our app @supermeme_ai in the future :) -
ruchowdh.bsky.social
@ruchowdh
Watching AI slowly discover the quantitative social sciences and claim it for its own is really a trip. People making whole careers poorly playing telephone with Gerber and Green, KKV, and others.
ruchowdh.bsky.social
@ruchowdh
Singleton and straits, Popper, Geertz.
Literally canonical, I learned them all in grad school. -
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
It's my great pleasure to share that I will be serving as an Ethics Review Chair at the 2023 @NeurIPSConf.
-
ruchowdh.bsky.social reposted
J.D. Capelouto
@jdcapelouto
.@ruchowdh on the pitfalls of governments embracing this new wave of AI technology:
www.semafor.com/article/03/03/2023/governments-using-chat… -
ruchowdh.bsky.social
@ruchowdh
This is the correct take on asking Kissinger anything on AI
gizmodo.com/chatgpt-ai-free-henry-kissinger-fake-news-wwi… -
ruchowdh.bsky.social reposted
Jeffrey Ladish
@JeffLadish
Well Meta's 65 billion parameter language model just got leaked to the public internet, that was fast.
Get ready for loads of personalized spam and phishing attempts. Open sourcing these models was a terrible idea -
ruchowdh.bsky.social
@ruchowdh
Agree with this thread - “art” is simply more than realistic replication or mimicking an existing style. There’s something humans pick up about honesty and earnestness. Being “realistic” speaks less to the animation than the lived experience.
-
ruchowdh.bsky.social reposted
Minderoo Centre for Technology and Democracy
@MCTDCambridge
"If platforms and policymakers do not address this issue now, countless unseen others will be silenced."
@ginasue and @ruchowdh write for @WIRED about the tools needed to fight chronic online abuse ⬇️
www.wired.com/story/platforms-combat-harassment-but-theyr… -
ruchowdh.bsky.social
@ruchowdh
Honored to be part of this global task force - emphasis on global - to guide the development of safe and secure algorithmic systems
Robert Muggah
@robmuggah
Predictive #ai is revolutionizing security and development around the world. While generating efficiencies they are also spreading new risks. @igarape_org and @NewAmerica are launching a global task force of 15 leading experts to ask hard questions. /1 igarape.org.br/en/digital-security/global-task-force/
-
ruchowdh.bsky.social
@ruchowdh
My op ed about the last days of Twitter is out. It was sad to reflect on all that could have been. I write this piece not just as a commemoration but as a warning - we need better systems of tech oversight www.theatlantic.com/technology/archive/2023/02/elon-musk-…
-
ruchowdh.bsky.social reposted
Damon Beres
@dlberes
"Months into Musk’s takeover—a new era defined by feverish cost-cutting, lax content moderation, the abandonment of important features... it seems no one is keeping watch" @ruchowdh shares her perspective from inside Twitter, and charts a path forward www.theatlantic.com/technology/archive/2023/02/elon-musk-…
-
ruchowdh.bsky.social reposted
Spherical Phil - Lawson
@SocializingAI
Context is ESSENTIAL but the #complexity of defining and understanding #context putting these variables into algorithms is challenging, not impossible. But it takes unique skills beyond coding
#wired on twitter research @peard33 @ruchowdh #ai #aiethics www.wired.com/story/the-scramble-to-save-twitters-researc… -
ruchowdh.bsky.social reposted
Billy Perrigo
@billyperrigo
Big tech companies once declined to release their most powerful AIs for safety reasons. Now, they are throwing open the doors.
Not because the AIs are any safer. But because a new arms race is on.
@TIME's cover story this week, by me and @andrewrchow
time.com/6255952/ai-impact-chatgpt-microsoft-google -
ruchowdh.bsky.social
@ruchowdh
::Cries in social scientist::
Molly White
@molly0xFFF
binance has produced a survey that found that buying crypto will totally help you get girls
-
ruchowdh.bsky.social
@ruchowdh
Watch Bing chat go slowly unhinged - @acidflask and I were messing around with it and asked it questions about me.
First response. Already factually incorrect currently but sort of true… but wait what’s that on the lower right?1/
ruchowdh.bsky.social
@ruchowdh
Oh uhhhh so that’s creepy but thanks I guess. Didn’t realize beauty could be “enlarged” and basic grammar is faltering. 2/
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
Finally this gem. I can’t comment on any Christian festivals of goddesses I’ve ever attended or movies I’ve ever been in.
Sounds like a nice life though - have $5m in the bank, act as a hobby and own 100 pairs of shoes while slowly enlarging my beauty.
ruchowdh.bsky.social
@ruchowdh
Thing is, I’m very online and there’s lots of information about me. But that information is almost entirely about my professional life. The fact that this generative bot can’t get basics about me correct and focuses more on very gendered speculation is deeply problematic.
-
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
“What is this new generation of AI good at, and where does it fall short?”
Listen to @BKCHarvard Responsible AI Fellow @ruchowdh discuss the state of generative AI with @scifri.
www.sciencefriday.com/segments/chatgpt-microsoft-google-a… -
ruchowdh.bsky.social reposted
Science Friday
@scifri
We're talking about how tech like #chatgpt could affect our lives this year with @MelMitchell1 and @ruchowdh on #scifrilive 📻
Have you used ChatGPT on work or at school? Are you worried about the ethics of AI?
Reply or ☎️ us at 1-844-SCI-TALK www.sciencefriday.com/segments/chatgpt-microsoft-google-a… -
ruchowdh.bsky.social reposted
moo
@moo_hax
Some terrible but functional Windows malware with ChatGPT. No jailbreaks needed. open.substack.com/pub/moohax/p/writing-malware-with-chatgpt
-
ruchowdh.bsky.social
@ruchowdh
If there’s one thing I’ve accomplished in life it’s closing out Sci Fri pointing out the flaws in extortive capitalism
Science Friday
@scifri
We're talking about how tech like #chatgpt could affect our lives this year with @MelMitchell1 and @ruchowdh on #scifrilive 📻
Have you used ChatGPT on work or at school? Are you worried about the ethics of AI?
Reply or ☎️ us at 1-844-SCI-TALK www.sciencefriday.com/segments/chatgpt-microsoft-google-a… -
ruchowdh.bsky.social
@ruchowdh
If you value your head, never tell the emperor he has no clothes
Rachel Cohen Booth
@rcobooth
i feel bad for this poor fired Twitter engineer but on the whole this story is so funny www.platformer.news/p/elon-musk-fires-a-top-twitter-engineer
-
ruchowdh.bsky.social
@ruchowdh
Yep it’s true! Melanie and I will be on @scifri tomorrow talking all things AI! Excited to be on the show again 🙏
Melanie Mitchell
@MelMitchell1
I'll be on public radio's Science Friday (@scifri) tomorrow (Feb 9) along with @ruchowdh, live at 2:20pm Eastern Time. We'll be discussing and answering audience call-in questions about generative AI.
-
ruchowdh.bsky.social
@ruchowdh
Trying to follow @JuliaConOrg and I can’t - I’m not even following 5000 people
-
ruchowdh.bsky.social
@ruchowdh
When Neil Gaiman comes for you it’s over
Neil Gaiman
@neilhimself
I just scheduled a Tweet for a minute from now. Twitter accepted it and then told me it would post on Feb 14th. Go home, Twitter. You're drunk. I wonder when this will post.
-
ruchowdh.bsky.social
@ruchowdh
I mean when Harvard Law School says you’re greedy that’s something else - papers.ssrn.com/sol3/papers.cfm?abstract_id=4330393
-
ruchowdh.bsky.social reposted
Sanjay Kairam
@skairam
Apropos of absolutely nothing, but here's a quick reminder that Reddit is an amazing place to study large-scale social communication and information-exchange, with data accessible for free to academics:
www.reddit.com/dev/api/ -
ruchowdh.bsky.social
@ruchowdh
Deeply interested in the tiktok ban discussion in congress because literally nothing better embodies the generational divide and the boomer stranglehold on political power
ruchowdh.bsky.social
@ruchowdh
And I don’t use the pejorative “boomer” lightly; was just watching a newscast and it was basically a bunch of 80 year old white men in suits talking about the harms of tiktok like they don’t get their emails printed out
-
ruchowdh.bsky.social
@ruchowdh
Strong “did you do the summer reading” vibes
Zoë Schiffer
@ZoeSchiffer
Elon Musk sent an email to Twitter employees telling them to write a one pager on what they accomplished in January, what they plan to accomplish in February, and "any ideas you have for improving Twitter." Twitter has also implemented another change freeze.
-
ruchowdh.bsky.social reposted
Mavi Doñate
@mavidonate
Muy interesante @ruchowdh directiva de META. Reconoce que en esta red que escribo, como en otras, las herramientas ante el acoso crónico son insuficientes. Parten de la filosofía de proteger, cuando el ataque está hecho, y no de evitar. El interés es cambiar esto dice
-
ruchowdh.bsky.social reposted
Nick Botton
@NickBotton
The study on digital advertising conducted by AWO on behalf of @EU_Commission was published today! Here’s a quick thread on its key findings. op.europa.eu/en/publication-detail/-/publication/8b950a43…
-
ruchowdh.bsky.social
@ruchowdh
Uh so I can’t seem to reply to tweets anymore
-
ruchowdh.bsky.social reposted
Oversight Board
@OversightBoard
Happening in 1⃣ hour!
Join @nighatdad, @AndrewPuddephat, and @ruchowdh as they discuss how we can create safer online spaces for all women and free from gendered #disinformation and gender bias while protecting freedom of expression.
Watch here: webcast.unesco.org/events/2023-01-IWDM/
Oversight Board
@OversightBoard
Tomorrow is International Day of #women in #multilateralism!
Join our Board Member @nighatdad tomorrow at @UNESCO for a discussion on the pressing global issue of social media content governance to counter gendered #disinformation.
You can watch the session here ↓ -
ruchowdh.bsky.social reposted
John Reed Stark
@JohnReedStark
Having served for 11 yrs as Chief of the SEC's Office of Internet Enforcement, I believe the design language of the crypto-Ponzi scheme is a multi-faceted grift, comprised of a wide range of falsehoods, all calculated to dupe consumers/investors. Here are 12 egregious examples:🧵
-
ruchowdh.bsky.social reposted
Elizabeth M. Renieris
@hackylawyER
Just received word that my forthcoming @mitpress book, Beyond Data, is 25% off at @bnbuzz with the code PREORDER25, today and tomorrow only.
Order here: www.barnesandnoble.com/w/?ean=9780262047821 -
ruchowdh.bsky.social
@ruchowdh
On a positive note - happy to announce that I’m joining @BKCHarvard to continue work on Responsible AI - in particular focusing on governance.
Berkman Klein Center for Internet & Society
@BKCHarvard
We are thrilled to announce that Dr. Rumman Chowdhury (@ruchowdh) has joined the annual @BKCHarvard fellowship cohort, working to strengthen and expand on the Center’s work on responsible AI.
cyber.harvard.edu/story/2023-01/announcing-dr-rumman-chow… -
ruchowdh.bsky.social
@ruchowdh
I’ve been keeping my account on private because of the significant rise in aggression on this platform, but coming back to public to share a few things, starting with this article -
Justin Hendrix
@justinhendrix
“It’s certainly very jarring. It’s very sad to see this thing that so many of us cared for and built being decimated piece by piece,” @ruchowdh told The Post. “It’s very hard to see where it’s headed and how bad it’s becoming.” www.washingtonpost.com/technology/2023/01/18/hate-speech-…
-
ruchowdh.bsky.social reposted
Justin Hendrix
@justinhendrix
“It’s certainly very jarring. It’s very sad to see this thing that so many of us cared for and built being decimated piece by piece,” @ruchowdh told The Post. “It’s very hard to see where it’s headed and how bad it’s becoming.” www.washingtonpost.com/technology/2023/01/18/hate-speech-…
-
ruchowdh.bsky.social reposted
Anna Kramer
@anna_c_kramer
w/ Google layoffs, a good day to remember that our immigration system is so screwed up that anyone laid off at Google on a visa (or Microsoft, or Amazon, etc) only has 60 days to get a job or get out of the US
www.wired.com/story/big-techs-layoffs-highlight-how-the-u… -
ruchowdh.bsky.social reposted
Lujain Ibrahim
@lujainmibrahim
After hearing @ruchowdh speak at the @oiioxford about AI bias & auditing, my classmates Andrew, Jonathan, and I decided to participate in her team’s Bias Bounty competition. We ended up getting the “Judge’s Commendation for Most Innovative Approach” www.technologyreview.com/2022/10/20/1061977/ai-bias-bount…
-
ruchowdh.bsky.social reposted
Billy Perrigo
@billyperrigo
🚨Exclusive: OpenAI used outsourced Kenyan workers earning less than $2 per hour to make ChatGPT less toxic, my investigation found
(Thread)
time.com/6247678/openai-chatgpt-kenya-workers/ -
ruchowdh.bsky.social reposted
lara mendonça 🇵🇸
@laraisuncool
Twitter is suggesting this topic to me, and when I click on it it shows me a bunch of photos and quotes of his speeches.
I don't think you all really understand how much radicalization Twitter is going to promote under Elon. -
ruchowdh.bsky.social reposted
Data Science Institute
@DSI_UChicago
Kristian Lum(@KLdivergence), a renowned statistician and machine learning researcher who focuses on fairness, accountability and transparency, has joined the Data Science Institute as a research associate professor.
Learn more about Kristian here: datascience.uchicago.edu/news/ml-bias-ethics-researcher-k… -
ruchowdh.bsky.social
@ruchowdh
Happy 2023, may this year be the year of integrity.
-
ruchowdh.bsky.social reposted
JC Oviedo
@JCOviedo6
"Based on the agreement Musk signed with bankers to help fund his Twitter purchase, he has already experienced his first margin call, an obligation to cough up cash or provide more collateral."
$ElonQ
www.barrons.com/articles/tesla-elon-musk-margin-call-twit… -
ruchowdh.bsky.social
@ruchowdh
If mastodon posts are being shared on Twitter does that mean twitter has fallen further in the social media cultural hierarchy? (Spot 1 held by tiktok of course)
M’s Sarfarosh, always
@BabaGlocal
This must hurt @elonmusk
-
ruchowdh.bsky.social
@ruchowdh
The only useful part about the views counter is that it’s a reminder that I should be on here less.
-
ruchowdh.bsky.social
@ruchowdh
I know I’m funny but how does one get 20 views and 62 likes?
-
ruchowdh.bsky.social
@ruchowdh
Guess that knowledge transfer is complete
Zoë Schiffer
@ZoeSchiffer
NEW: Luke Simon -- one of Elon Musk's top lieutenants at Twitter -- appears to be out. He's been deactivated on Slack.
-
ruchowdh.bsky.social
@ruchowdh
What prize do I win when when tiktok fyp shows me @igb going at Musk? (This was glorious, kudos)
www.tiktok.com/t/ZTRqr7D3W/ -
ruchowdh.bsky.social reposted
Jason Goldman
@goldman
Listening to this Twitter Spaces where Elon is talking about the tech stack and it's clear that no one has been in a position to actually ask him a follow up question in years. He thinks he knows everything and simply will not abide someone contradicting that notion.
-
ruchowdh.bsky.social reposted
RΛISINI ライシニ
@raisini
Thread 🧵
Elon Musk is a pathological liar, a charlatan and a Super Villain. I have nothing against him but as an investor in Tesla & Twitter, i have serious concerns about his long track record of lies, his treatment of staff at Tesla, SpaceX & Twitter. This thread is about that -
ruchowdh.bsky.social reposted
Andrey Kurenkov
@andrey_kurenkov
Should @elonmusk shut down Twitter? He will abide by the results of this poll.
-
ruchowdh.bsky.social
@ruchowdh
Tell me you don’t understand social media without telling me you don’t understand social media
-
ruchowdh.bsky.social
@ruchowdh
Wow two whole dollars
Judd Legum
@JuddLegum
IMPORTANT: @elonmusk's new policy preventing linking to competitors is in DIRECT VIOLATION of European Union rules.
Penalty can be a fine of up to 20% of Twitter's annual revenue
commission.europa.eu/strategy-and-policy/priorities-2019-… -
ruchowdh.bsky.social reposted
Judd Legum
@JuddLegum
IMPORTANT: @elonmusk's new policy preventing linking to competitors is in DIRECT VIOLATION of European Union rules.
Penalty can be a fine of up to 20% of Twitter's annual revenue
commission.europa.eu/strategy-and-policy/priorities-2019-… -
ruchowdh.bsky.social
@ruchowdh
Ah the PayPal mafia
Tech Won't Save Us
@techwontsaveus
🚨 THE FINAL TWO 🚨
Tens of thousands of voters narrowed 32 candidates down to just two. Today, the PayPal rivals and right-wing billionaires harming the world face off.
It’s Peter Thiel vs Elon Musk.
Cast your vote to choose who takes the title of Worst Person in Tech. -
ruchowdh.bsky.social
@ruchowdh
Bro not to humblebrag but I knew john mastodon in college
-
ruchowdh.bsky.social reposted
Sal Gentile
@salgentile
hard to believe there was once a time where they’d send a big yellow book to everyone’s house listing yours and everyone else’s assassination coordinates
-
ruchowdh.bsky.social reposted
Věra Jourová
@VeraJourova
News about arbitrary suspension of journalists on Twitter is worrying. EU’s Digital Services Act requires respect of media freedom and fundamental rights. This is reinforced under our #mediafreedomact. @elonmusk should be aware of that. There are red lines. And sanctions, soon.
-
ruchowdh.bsky.social reposted
Elizabeth M. Renieris
@hackylawyER
I don’t see how moving to something funded by a16z is good alternative 🤷♀️
-
ruchowdh.bsky.social reposted
Shannon Liss-Riordan
@SLissRiordan
This decision is a victory for Twitter workers who have been abused by @ElonMusk. @Twitter must notify employees of our legal action, giving them the chance to more fully understand their rights instead of just signing them away under pressure from Musk. www.bloomberg.com/news/articles/2022-12-14/twitter-must-t…
-
ruchowdh.bsky.social
@ruchowdh
Wow it really took this blubbering infant a month before he imploded. So much for free speech, but we already knew this.
-
ruchowdh.bsky.social reposted
Kat Tenbarge is on Bluesky
@kattenbarge
Elon Musk was just in a Twitter Space with a few journalists and after getting like 3 questions he had to leave because he couldn’t answer a single one
-
ruchowdh.bsky.social reposted
@eliothiggins.bsky.social
@EliotHiggins
Bellingcat Discord member Sepulco geolocated the video Elon Musk posted overnight, claiming to show a stalker, which he used to justify the banning of accounts that tracked flights. It's at 34.116175,-118.160635, not close to any airports.
www.google.com/maps/@34.1161302,-118.1607175,3a,75y,63.89… -
ruchowdh.bsky.social reposted
.
@naledimashishi
People aren’t abusers because they don’t have access to therapy and therapy often gives abusers tools to abuse their victims in new and inventive ways. They’re abusers because they believe they are entitled to exert control over their victims
-
ruchowdh.bsky.social
@ruchowdh
So. Who’s writing the story about the ww who have enabled Elon - Shotwell, Crawford, Irwin, etc.? He’s only able to be a 24/7 asshole and do no work because they enable him.
-
ruchowdh.bsky.social
@ruchowdh
Yes they’re all shit people
-
ruchowdh.bsky.social reposted
A. Derfel
@a_derfelGazette
1) More and more reports are emerging of Twitter's security vulnerabilities since @elonmusk's takeover of the social media platform a few weeks ago, as well as efforts to hack the accounts of certified journalists. In this thread, I will survey how unsafe Twitter has become.
-
ruchowdh.bsky.social reposted
Dan Primack
@danprimack
No matter if you think @yoyoel did an amazing or terrible job at Twitter, or have no opinion at all, what @elonmusk did to him yesterday was both dangerous and disgusting.
It’s the sort of thing that would cause most boards to call a special meeting. But Twitter doesn’t have one -
ruchowdh.bsky.social reposted
A. Derfel
@a_derfelGazette
1) Hi, everyone. I wanted to share with you the scary recent experience I had with my original Twitter account, @Aaron_Derfel, which was hacked two days ago, and how I fear this raises serious concerns about the future of this social media platform.
-
ruchowdh.bsky.social reposted
Tom Coates
@tomcoates
This is Elon Musk attempting to misrepresent his former head of Trust and Safety, who he lauded only a couple of weeks ago, as somehow a threat to children or some kind of predator to his 120,000,000 million followers.
-
ruchowdh.bsky.social
@ruchowdh
Twitter 2.0 is now erasing my existence? This blog post no longer has my name on it. I seem to be on the other ones.
blog.twitter.com/en_us/topics/company/2021/rml-politicalc… -
ruchowdh.bsky.social reposted
parker lyons
@tweetsbyparker
fair and balanced
-
ruchowdh.bsky.social
@ruchowdh
It is laughable and yet depressing to see that elons Twitter is so broke they’re selling premium furniture and espresso machines for $25
-
ruchowdh.bsky.social reposted
bonappetweet
@bonappetweet
Hi, Twitter Fam just wanted to say our final good byes. It’s been a pleasure serving you this past 11 years. Twitter really made us feel apart of the culture and we will never forget that. 🎤🎤 drop
-
ruchowdh.bsky.social
@ruchowdh
Still can’t lock my account. Cool cool everything’s fine on the bird app
ruchowdh.bsky.social
@ruchowdh
Putting this account on lock and will only be here moving forward to post general updates.
You know, kinda like how people treat Facebook.
Find me on Mastodon if you want to talk ♥️ -
ruchowdh.bsky.social
@ruchowdh
Putting this account on lock and will only be here moving forward to post general updates.
You know, kinda like how people treat Facebook.
Find me on Mastodon if you want to talk ♥️
ruchowdh.bsky.social
@ruchowdh
Wow well I’d like to but I guess twitters broken so I can’t?
-
Dr. Sarah T. Roberts ubiquity75@mastodon.social
@ubiquity75
Hey, @ellagirwin — can you explain why “egregious spam” is a sacrosanct category? Please tell the truth, that it’s much easier to automate than moderation requiring a brain. Can you stop lying about who and how many people were fired from Health? www.reuters.com/technology/twitter-exec-says-moving-fast-…
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
What I do recall from our overlapping time is the constant dissatisfaction of talented people and the brain drain BEFORE Elon took over. Friends who were reorged until their heads spun and they lost faith.
-
ruchowdh.bsky.social
@ruchowdh
The sexual harassment to conservative pundit pipeline
-
ruchowdh.bsky.social reposted
Dr. Sarah T. Roberts ubiquity75@mastodon.social
@ubiquity75
Hey, @ellagirwin — can you explain why “egregious spam” is a sacrosanct category? Please tell the truth, that it’s much easier to automate than moderation requiring a brain. Can you stop lying about who and how many people were fired from Health? www.reuters.com/technology/twitter-exec-says-moving-fast-…
-
ruchowdh.bsky.social
@ruchowdh
Wow speculative garbage mixed with unverifiable contextless “screenshots” presented by a complete hack so fucking compelling I’m totally redpilled now
-
ruchowdh.bsky.social reposted
X Daily News
@xDaily
NEWS: Twitter is offering aggressive incentives to get advertisers on the platform before the end of the year.
-
akivamcohen.bsky.social
@AkivaMCohen
So here's the thing. You can only violate people's legal rights and your own word so far before they lawyer up and come after you.
I really do hope Musk changes his mind and does the right thing - the employees deserve that. But it'll be fun as hell if he doesn't.
ruchowdh.bsky.social
@ruchowdh
It’s signed “all the best” so you know he’s going to mess someone up
-
ruchowdh.bsky.social reposted
akivamcohen.bsky.social
@AkivaMCohen
So here's the thing. You can only violate people's legal rights and your own word so far before they lawyer up and come after you.
I really do hope Musk changes his mind and does the right thing - the employees deserve that. But it'll be fun as hell if he doesn't. -
ruchowdh.bsky.social reposted
Ferenc Huszár
@fhuszar
Hey @realGeorgeHotz. Lots of people are requesting access to the data from www.pnas.org/doi/10.1073/pnas.2025334119 so they can verify and scrutinize these claims.
None of us authors work there anymore. Can you make this open access? -
ruchowdh.bsky.social reposted
Mike Drucker
@MikeDrucker
I feel like we’re close to cracking the mystery of whether the man who grew up wealthy from Apartheid might be racist
-
ruchowdh.bsky.social reposted
Brianna Wu
@BriannaWu
4/ Don’t let him trick you with “but muh free speech.” He’s been redpilled, and wants no rules around racism, antisemitism, transphobia and inciting terrorism for a fringe right political agenda.
Decent people can express themselves without making death threats. -
ruchowdh.bsky.social reposted
Brianna Wu
@BriannaWu
🧵 Remember: Apple’s App Store rules were VERY SPECIFICALLY DESIGNED to stop an Elon Musk. A rich, giant corporation that COULD do the right thing for users, but decided it would be cheaper and easier not to.
These rules protect your children from being served porn, for instance -
ruchowdh.bsky.social
@ruchowdh
Dense and intense is exactly how I’d describe Elon
Zoë Schiffer
@ZoeSchiffer
NEW: Email from Elon to Twitter engineers and designers: "Anyone writing software or doing design should be on the 10th floor of SF HQ. It is intended to be dense and intense. Thanks, Elon."
-
ruchowdh.bsky.social reposted
X Daily News
@xDaily
CONTEXT: Apple spent 50 million dollars in advertising on $TWTR IN Q1.
Apple pulling out is a big blow to Twitters revenue. -
ruchowdh.bsky.social reposted
Zoë Schiffer
@ZoeSchiffer
NEW: Twitter engineers have received another late night email laying out Elon Musk’s expectations. Managers are expected to “identify low performers on a regular basis.” Those employees then have “up to” 4-weeks to improve or they’re out. 1/
-
ruchowdh.bsky.social reposted
sasha solomon
@sachee
My bedside table
-
ruchowdh.bsky.social reposted
Chris Stokel-Walker
@stokel
SCOOP: Twitter didn't pay its current and former employees on time in the UK and Germany today. “It has come to our attention that some of you may not have received your November 2022 salary yet in your bank account,” an email sent this morning says.
-
ruchowdh.bsky.social reposted
Christopher Bouzy (spoutible.com/cbouzy)
@cbouzy
When I suggested Apple was going to pull Twitter from the App Store, a lot of people said that would never happen. A lot of people underestimated Apple.
-
ruchowdh.bsky.social reposted
Jane Manchun Wong
@wongmjane
I have evidences that some former Twitter employees are actually time travelers
They traveled to the future to steal the great inventions from Twitter 2.0 -
ruchowdh.bsky.social
@ruchowdh
Question: what’s the accountability here? He lies and we know it so…
Best of Dying Twiter
@bestofdyingtwit
-
ruchowdh.bsky.social reposted
Am I the Asshole?
@AITA_online
It's an honor just to be nominated
-
ruchowdh.bsky.social reposted
-
ruchowdh.bsky.social
@ruchowdh
Elon lies. A few days ago, he told the company that there were no more layoffs planned.
I’m sure he has a weaselly excuse about how these pre thanksgiving firings are “performance related.”
Bullshit. -
ruchowdh.bsky.social
@ruchowdh
What’s fucking wild about Apartheid Clyde is that he seems all about the letter of the law when it involves black ppl but not when it involves compliant companies 🤔 must be all that genius
-
ruchowdh.bsky.social reposted
Best of Dying Twiter
@bestofdyingtwit
wait WHAT
-
ruchowdh.bsky.social reposted
-
ruchowdh.bsky.social reposted
Jane Manchun Wong
@wongmjane
Some people left for Mastodon
Some people left for Hive
Some people returned to IG / Tumblr
Some people just left -
ruchowdh.bsky.social
@ruchowdh
This is just depressing at this point.
Casey Newton
@CaseyNewton
More layoffs hit Twitter’s sales team this morning, per someone affected. Account managers and client partners started to be informed last night.
Comes a day after Musk held a short-notice Sunday evening all-hands meeting with the sales team where layoffs were not discussed. -
ruchowdh.bsky.social reposted
Peter Yang
@petergyang
An advertiser explains why they’re pausing their Twitter ads campaigns:
-
ruchowdh.bsky.social
@ruchowdh
I’ve received more convincingly written spam emails btw
kate conger
@kateconger
In other news, Twitter sent a note to laid off employees tonight letting them know that their separation agreements will be further delayed.
-
ruchowdh.bsky.social reposted
Alf
@yun
This is the epitome of Twitter 1.0 and no amount of money or hubris will ever be able to destroy it. The story behind the #lovewhereyouwork hashtag.
💙 Tweeps, our bond is unbreakable. 💙
#lovewhereyouworked #oneteamThis media couldn't be displayed. See it on the original post. -
ruchowdh.bsky.social reposted
Brianna Wu
@BriannaWu
This is amazing. Elno broke the copyright strike system, so you can just post entire films on Twitter now. 😂
-
ruchowdh.bsky.social reposted
Shannon Liss-Riordan
@SLissRiordan
BREAKING: Court agrees @Twitter cannot seek employee releases until decision on Emergency Motion! This is the first in what we hope will be a series of victories for workers who were mistreated by @ElonMusk. See you in Court next month, Elon.
Shannon Liss-Riordan
@SLissRiordan
This emergency motion is critical to protecting @Twitter employees’ rights. @ElonMusk is trying his hardest to screw over his workers, but we are prepared to fight him and get the employees what they are owed under the law.
-
ruchowdh.bsky.social reposted
Dave Lee
@DaveLeeBBG
Journalists moving from Twitter to Mastodon tonight — we’ve built a simple service to verify and share journalists’ profiles across *any* Mastodon server.
Verified accounts get a badge to signal authenticity.
Apply here: www.presscheck.org -
ruchowdh.bsky.social reposted
David Beard
@dabeard
Hey, thought everyone should be able to read this story about the former head of Twitter's trust and safety thinks will happen. So here's a free version. www.nytimes.com/2022/11/18/opinion/twitter-yoel-roth-elon…
-
ruchowdh.bsky.social
@ruchowdh
If anyone still at Twitter is going to this stupid “code review” by El Musko I will pay you cash money to walk in there with a document in binary and present in earnest about the hardcoreness of your code.
ruchowdh.bsky.social
@ruchowdh
I’m getting offers to pitch into this fund, does this mean I have a startup
ruchowdh.bsky.social
@ruchowdh
Can Elon and I have a race to the bottom of how few engineers we need
-
ruchowdh.bsky.social
@ruchowdh
Elon employees buying peroxide like son of Sam 2022
-
ruchowdh.bsky.social reposted
Ramin Khatibi
@Ramin_DK
It's time for a thread on why Twitter hasn't gone down yet 🧵
-
ruchowdh.bsky.social
@ruchowdh
ELL OH ELL
-
ruchowdh.bsky.social
@ruchowdh
In case it wasn’t evident, my twitter account is now an Elon shitposting account and all my real content has moved to Mastodon. Find me at ruchowdh@mastodon.social
-
ruchowdh.bsky.social reposted
Laura Gao ☕️
@heylauragao
Twitter employees take pride in being able to speak up to power, the same principle this platform was built on. 2 yrs ago, I openly criticized execs for their lack of diversity. Instead of firing me, they reached out directly and ran focus groups with women and folks of color.
Laura Gao ☕️
@heylauragao
Harvard Professor: “Now Tweeps, what is wrong with this photo?”
-
ruchowdh.bsky.social reposted
Berkman Klein Center for Internet & Society
@BKCHarvard
NEW: @BKCHarvard is also accepting Visiting Scholar applications for its Institute for Rebooting Social Media 2023-2024 cohort. This call is specifically for faculty members committed to improving the current state of social media.
Deadline: Dec. 9, 2022
rebootingsocialmedia.org/call-for-2023-2024-rsm-visiting-… -
ruchowdh.bsky.social
@ruchowdh
Lol so that email really is a grade school note
Do you like me?
✅ yes
✅ no
(Tweeps know what I’m talking bout - if you don’t you will soon) -
ruchowdh.bsky.social reposted
-
ruchowdh.bsky.social reposted
Oxford Internet Institute
@oiioxford
Delighted to have @ruchowdh pop in to the OII today to talk candidly about algorithmic bias, tech layoffs, her new projects, Twitter & #elon (obvs) and much more.
Thanks so much, Rumman. -
ruchowdh.bsky.social reposted
Melissa McCradden
@MMccradden
So much respect for the Twitter META team @ruchowdh nurtured 🙏🏻 and sorry to see what’s happening now. Your work matters and is appreciated by so many more than you know. #twitterlayoffs
-
ruchowdh.bsky.social
@ruchowdh
This should be everyone’s 2023 vibe
soul nate
@MNateShyamalan
can’t believe my stupid ass has imposter syndrome while elon musk is confidently marching around twitter HQ and chewing on wires until something breaks
-
ruchowdh.bsky.social reposted
Dr. Sarah T. Roberts ubiquity75@mastodon.social
@ubiquity75
Elon like: “Hey, sorry about latency issues from the RPC calls. XTF runtime protocols involved for client-side rendering of home GER are being affected by the STI VPD that I luckily caught in the code printouts.
I also fired everyone who worked here. And am not an engineer.” -
ruchowdh.bsky.social reposted
Marlow Stern
@MarlowNYC
why would anyone want to work for this person
-
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
ELI5, Andreas. Why is it okay for you to be divisive (literally tag and call someone out for being ‘bitter’ etc) but not OK for her to do the same?
Let’s just say someone agrees with your statement. Why are you exempt from your own judgment? -
ruchowdh.bsky.social reposted
ACM FAccT
@FAccTConference
We're so excited about next year's #facct23 Conference! Taking place in Chicago, in mid June, the General Chairs are Aziz Huq and Chenhao Tan of U Chicago, and Christina Harrington and Sarah Fox of CMU. @adapperprof @ChenhaoTan
For updates see facctconference.org/2023/index.html -
ruchowdh.bsky.social reposted
Ferenc Huszár
@fhuszar
I wouldn’t trade our meaningful algorithmic transparency research for Yann’s meaningful AI research™️
Yann LeCun
@ylecun
I suppose things changed for the better when @mmbronstein was acquihired.
But that's a relatively small outfit.
There were previous failed attempt to convince Twitter to have a meaningful AI research efforts, e.g. by @ryan_p_adams @hugo_larochelle @clmt and others. -
ruchowdh.bsky.social reposted
Casey Newton
@CaseyNewton
Update: company sources tell me that yesterday Twitter eliminated ~4,400 of its ~5,500 contract employees, with cuts expected to have significant impact to content moderation and the core infrastructure services that keep the site up and running.
People inside are stunned. -
ruchowdh.bsky.social reposted
ruchowdh.bsky.social
@ruchowdh
At this point we are all just egging elons house.
-
ruchowdh.bsky.social reposted
Jason Goldman
@goldman
I don’t think anyone routinely reveals themselves to be as profoundly incurious about how the web has evolved in the last 20 years as the owner of this website
-
ruchowdh.bsky.social
@ruchowdh
If any former Twitter contractor needs a community of support pls reach out. There’s a lot of ex Tweep groups and that means you too 🫡
-
ruchowdh.bsky.social reposted
Rhodes Trust
@rhodes_trust
Thanks to @hare_brain, & @ruchowdh for opening #techandsociety2022 with a discussion of technology #ethics in action. And a huge thank you to everyone joining us in person and virtually! Join the conversation using #techandsociety2022💡
-
ruchowdh.bsky.social
@ruchowdh
Between crypto and Elon We all deserve Monday off to just lay in bed and make fun of these people on Twitter
-
ruchowdh.bsky.social reposted
rat king 🐀
@MikeIsaac
here is a story on the past two weeks inside Elon Musk’s takeover of twitter
w/ @kateconger @RMac18 @tiffkhsu
www.nytimes.com/2022/11/11/technology/elon-musk-twitter-t… -
ruchowdh.bsky.social
@ruchowdh
Hear me out - what if we gamified twitter
Zack Kanter
@zackkanter
instead of a subscription you should have to buy checkmarks like lives on twitter and when you dunk hard enough on someone you get their checkmark and they are dead until they buy more
-
ruchowdh.bsky.social reposted
Dolly Madison ✌🏻🌻 🟦
@dollymad1812
-
ruchowdh.bsky.social reposted
Ben Smith
@semaforben
NPR issues internal guidance on Twitter: don't advertise your handle, but don't delete it because someone could take it and impersonate you.
-
ruchowdh.bsky.social reposted
Billy Perrigo
@billyperrigo
New: TikTok outsourced traumatic content moderation work to Colombia. Now, following an exposé on low pay + alleged union busting there by @TIME and @TBIJ, the Colombian government is officially investigating TikTok's subcontractor.
time.com/6231625/tiktok-teleperformance-colombia-investig… -
ruchowdh.bsky.social reposted
Ben Rossen
@benrossen
This is a really excellent question!
-
ruchowdh.bsky.social
@ruchowdh
Mark Cuban with the erudite contributions to the twitter debacle was def not in my bingo card
-
ruchowdh.bsky.social
@ruchowdh
Seeing this too
-
ruchowdh.bsky.social reposted
MMitchell
@mmitchell_ai
Former METAmates and FAIRies!
Did you work on AI Safety* features like watermarking, data consent, explicit content? Please apply 👇
apply.workable.com/huggingface/j/8F1299F490/
*Not the Effective Altruism-laden-AI-Safety. I'm hesitant to use the "AI Safety" term. -
ruchowdh.bsky.social reposted
Linette Lopez
@lopezlinette
How do I know so much about how @elonmusk does things? I spent 3 years investigating Tesla at @BusinessInsider from 2018-2021. Here are some of sloppy, dangerous, callous, things I learned:
-
ruchowdh.bsky.social reposted
Jack Clark
@jackclarkSF
Shoutout to @ruchowdh and the rest of the team for doing some of the most interesting applied work on AI ethics. I really enjoyed their investigation of the Twitter cropping algo blog.twitter.com/engineering/en_us/topics/insights/2021/s…
-
ruchowdh.bsky.social
@ruchowdh
It’s beautiful that this work came out as Twitter implodes. This was the last thing I worked on as Parity CEO before making the move. Congratulations on a phenomenal project.
Center for Critical Race and Digital Studies
@centerforcrds
Proud to announce: "A People’s Guide to Finding Algorithmic Discrimination." Funded by @DemocracyFund this practical guide to understanding this topic was led by CR+DS affiliates @merbroussard @kingavriel in partnership with @ruchowdh & @ParityAI www.criticalracedigitalstudies.com/peoplesguide
-
ruchowdh.bsky.social reposted
MMitchell
@mmitchell_ai
Former Tweeps!
Did you work on AI Safety* features like watermarking, data consent, explicit content? Please apply 👇
apply.workable.com/huggingface/j/8F1299F490/
*Not the Effective Altruism-laden-AI-Safety. I'm hesitant to use the "AI Safety" term. -
ruchowdh.bsky.social reposted
𝙷𝚒𝚖𝚊 𝙻𝚊𝚔𝚔𝚊𝚛𝚊𝚓𝚞
@hima_lakkaraju
Excited to share that my day-long workshop (a short course) on #explainableai is now publicly available as a five-part youtube video lecture series.
Link to video lectures: www.youtube.com/playlist?list=PLoROMvodv4rPh6wa6PGcHH6vMG…
Link to slides: docs.google.com/presentation/d/1khY_li29A5aUo_cEVRsvO8pcR…
#ai #ml @trustworthy_ml @XAI_Research -
ruchowdh.bsky.social
@ruchowdh
Elon jibes aside (and they are hilarious) it’s amazing how people like him don’t know what it’s like on a daily basis as a minority on this platform. Your feelings hurt lil buddy?
We don’t get jokes, we get death threats and rape threats. -
ruchowdh.bsky.social reposted
Alex Gwyn
@AlexKGwyn
It's only Birdwatch if it's from the Twittèr region of France, otherwise it's just sparkling community notes
-
ruchowdh.bsky.social reposted
Susan Lee Neth
@soozinbird
-
ruchowdh.bsky.social reposted
Hoodlum 🇺🇸
@NotHoodlum
General Mills, CVS, United Airlines, GM, Audi, Mazda, Porsche, VW, American Express, Coca-Cola, Johnson & Johnson, Levi Strauss, Spotify, Ford, Dyson, Forbes, DIRECTV, Nintendo, Unilever, and PBS have all suspended ads on Twitter.
-
ruchowdh.bsky.social
@ruchowdh
I’ve been waiting for my notifications to die down a bit before posting again because virality is overwhelming.
Thank you all for the outpouring of support and positivity to Twitter’s former ML ethics team.
I’ll have more to say later - esp after Jan 4 😉 -
ruchowdh.bsky.social
@ruchowdh
Petition to have @tweetsbyparker guest write for SNL
Saturday Night Live
@nbcsnl
a look at the twitter content moderation council meeting
This media couldn't be displayed. See it on the original post. -
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Pushed to prod without copyediting huh
-
ruchowdh.bsky.social reposted
parker lyons
@tweetsbyparker
-
ruchowdh.bsky.social reposted
Distributed AI Research Institute is on Mastodon
@DAIRInstitute
Twitter's machine learning ethics & transparency (META) team led by our friend @ruchowdh, has been laid off along with countless other workers. What's happening there should be a wakeup call on the consequences of leaving our technological future to the whims of billionaires.
-
ruchowdh.bsky.social reposted
Dr. Sarah T. Roberts ubiquity75@mastodon.social
@ubiquity75
Does this mean, @jack, that you feel the specific ways in which the “company grew too fast” was in the form of teams like META (ethical AI), Human Rights, Curation (working to stop the flow of global election-related misinformation)?
Curious as to your take. -
ruchowdh.bsky.social reposted
@timnitGebru (@dair-community.social/bsky.social)
@timnitGebru
“I hired too many of those people working on accessibility, human rights & ethics too fast and the manchild who wants to extend the light of consciousness got rid of them and he’s right.”
Can we abolish billionaires yet? -
ruchowdh.bsky.social reposted
𝙽𝚒𝚌𝚔 𝙼𝚊𝚝𝚑𝚎𝚜𝚘𝚗
@willplayforfood
So many feels seeing the praise of the Twitter META team as this chapter comes to a close
I want to particularly recognize @naz_erkan who conspired with @__lucab to start the whole thing
Thank you Naz for all the unseen labor that made space for this team to ultimately flourish -
ruchowdh.bsky.social reposted
MMitchell
@mmitchell_ai
All my love to @ruchowdh @KLdivergence and others at Twitter who were doing the hard work of ethics in tech. Appreciate you. 💜💜 gizmodo.com/twitter-layoffs-elon-musk-ai-ethics-1849743051
-
ruchowdh.bsky.social
@ruchowdh
I keep thinking about last night.
It was the most humanity-affirming moment that as each tweep was fired, we all posted a 🫡. Then we laughed and rejoiced in the decency and kindness of friends.
What a sendoff, Twitter. -
ruchowdh.bsky.social reposted
Alex Hanna (اليكس حنٌا)
@alexhanna
A word about the META team build by @ruchowdh @quicola and others --
This was probably one of the last teams at a big tech/social company that had the ear of product and policy, that wasn't dismantled at the whim of a whiny white tech boy. 1/ -
ruchowdh.bsky.social reposted
Seth Lazar
@sethlazar
So grateful to @ruchowdh and Kristian Lum for all their work for FAccT22 (and in Kristian’s case many before that). So impressed by their work, and that of @binary_aaron @__lucab and many other META tweeps who came to Seoul and made such a big difference. Onwards, comrades ❤️
Jacob Metcalf 🐀
@undersequoias
My tweep friends, especially on the META team, we saw you and were always impressed with your intelligence and integrity. 🫡
-
ruchowdh.bsky.social
@ruchowdh
JORTS REPLIED! It’s definitely only downhill on this hellsite.
Jorts (and Jean)
@JortsTheCat
I have a question: where do you live
This post is unavailable. See it on X. -
ruchowdh.bsky.social reposted
Jacob Metcalf 🐀
@undersequoias
My tweep friends, especially on the META team, we saw you and were always impressed with your intelligence and integrity. 🫡
-
ruchowdh.bsky.social reposted
holyn
@holyn
tweeps, your former colleagues (and others) are here for you
i’ve created this list of people who’ve tweeted they want to help whether that’s resume reviews, referrals, chats, etc. and i’ll continue to add to it
x.com/i/lists/1588345609815314432 -
ruchowdh.bsky.social
@ruchowdh
What a bittersweet phrase - not because I’m gone, but because it’s gone.
#lovewhereyouworked -
ruchowdh.bsky.social
@ruchowdh
🫡
-
ruchowdh.bsky.social
@ruchowdh
Imagine feeling relief that you’re the one on the receiving end of the thanos snap 🫰
-
ruchowdh.bsky.social
@ruchowdh
Has it already started? Happy layoff eve!
-
ruchowdh.bsky.social
@ruchowdh
I don’t know or care if this is true but it’s now canon
emily
@emcorinne
the fact that the first thing tweeps did when layoffs were finally, officially announced was to send so many heart emojis that we broke slack says everything no one else knows about this place. #oneteam
-
ruchowdh.bsky.social
@ruchowdh
I’m so glad joan shared this because it needs to be out in the world ❤️
-
ruchowdh.bsky.social
@ruchowdh
I’m so glad joan shared this because it needs to be out in the world ❤️
-
ruchowdh.bsky.social
@ruchowdh
I guess corporations are people but I’ve never had one send me an email before
-
ruchowdh.bsky.social
@ruchowdh
A thoughtful guide for Tweeps. Stay smart, not scared.
Collective Action in Tech
@tech_actions
Last night, friends from the birdapp reached out to us, asking us to publish a layoff guide for their coworkers.
It is written thoughtfully and with great care. If you work at the company, we encourage you to take a look.
collectiveaction.tech/2022/a-layoff-guide-for-tweeps/ -
ruchowdh.bsky.social reposted
Sabrina
@NerdyAndQuirky
billionares are so weird. what are you saving up for? hell?
-
ruchowdh.bsky.social
@ruchowdh
-
ruchowdh.bsky.social
@ruchowdh
-
ruchowdh.bsky.social reposted
BiasBounty
@BiasBounty
Thee winners o’ o’r pirate photo competition:
First: @acidflask
Second: @neilturkewitz
Third: @comathematician
Stay tuned fer yr prizes, mates, and keep yr DMs open for details!
And keep yer eyes peeled fer another treasure hunt! -
ruchowdh.bsky.social reposted
Mostly here now: @davidthewid.bsky.social
@davidthewid
Today in our Critical AI studies group at 3 ET, @struthious will be leading discussion of "Where Responsible AI Meets Reality", a paper I really enjoyed by @bobirakova @jingyingyang, @hsmcramer, @ruchowdh.
Anyone is welcome, DM me your email to be added to the calendar invite!
Mostly here now: @davidthewid.bsky.social
@davidthewid
Planning to co-lead a Critical AI studies reading group, for people inside and outside of the academy/CS/AI/CMU.
Interested? 1st meeting early July, Fill in this when2meet w your weekly availability www.when2meet.com?15918524-tB4t4, and reply and reading recs you might have! -
ruchowdh.bsky.social
@ruchowdh
It’s almost Monday - stay smart, not scared, tweeps 💙
-
ruchowdh.bsky.social reposted
Haben Girma👩🏿🦯
@HabenGirma
How do blind people use #mastodon? I tried it and the Compose button does not work with VoiceOver (the screenreader blind people use on iOS). @joinmastodon #twittermigration #mastodonsocial #a11y
-
ruchowdh.bsky.social
@ruchowdh
Word of the day: integrity.
-
ruchowdh.bsky.social reposted
Michelle
@_chellele
a lot of people who never worked at twitter sure have a lot of opinions about what it’s like working at @Twitter
-
ruchowdh.bsky.social reposted
Mastodon (@Mastodon@mastodon.social)
@joinmastodon
Signing up on Mastodon is easy! Unlike a traditional social network, it's not a single website. Instead, you pick a "server" which is what hosts all your data, similar to an e-mail provider. Each server gives you access to the same social network.
-
ruchowdh.bsky.social reposted
Crystal Lee (@crystaljjlee on all platforms)
@crystaljjlee
Twitter friends, if you’re on Mastodon, could you drop your handle here? Would love to follow what you say there, too! Recs for other folks to follow also very welcome!
-
ruchowdh.bsky.social reposted
wells (wants a better world)
@WellsLucasSanto
In case you're leaving Twitter for Mastodon, we're using the #twittermigration hashtag there to introduce ourselves there! Spread the word!
mastodon.social -
ruchowdh.bsky.social reposted
Aaron
@Golisms
This is a great idea. We need to do a better job understanding both neural networks and the underlying data. For anyone who can’t get through the firewall, here’s the link to the competition: biasbounty.ai
-
ruchowdh.bsky.social
@ruchowdh
What’s everyone drinking for all hands
-
ruchowdh.bsky.social reposted
Mastodon (@Mastodon@mastodon.social)
@joinmastodon
Looks like #mastodon is trending on Twitter as more and more people are announcing their new profiles. Welcome to the better social media that does not belong to a single company and cannot be sold, welcome to the fediverse!
-
ruchowdh.bsky.social
@ruchowdh
If you are this student PLEASE DO THIS you will make your former profs year
todd dillard
@toddedillard
professors, if a student you haven't spoken to/interacted with in 15ish years wrote to you and said "hey you were a great teacher I am still benefitting from your teaching, here's a small life update" would that be weird? would you assume the former student wanted something?
-
ruchowdh.bsky.social reposted
Martin
@hutko
The final text of the Digital Services Act was published this morning eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ:L:2022… Get used to Regulation (EU) 2022/2065!
-
ruchowdh.bsky.social reposted
Charlie Warzel
@cwarzel
a lot of the worst/angriest posters have made the place undesirable for a lot of ppl I loved to hear from. now they mostly lurk or have left. i think about how that makes people feel when *they* post bc the experience is awful when you're only gonna hear from the loud/angry ones
Kenneth Li
@kenli729
EXCLUSIVE Twitter is losing its most active users, internal documents show www.reuters.com/technology/exclusive-where-did-tweeters-g… by @Sheila_Dang
-
ruchowdh.bsky.social
@ruchowdh
I was listening to a murder podcast like a basic bitch and the hosts said they dropped twitter once they heard Elon was taking over THERE TRULY IS NO ESCAPE can I hear about homicide in peace
Robin Wood
@robonthestreet
I’m at trivia and one of the questions was when does the Twitter Deal close THERE IS NO ESCAPE
-
ruchowdh.bsky.social reposted
kat telles
@kltelles
The five stages of grief for Tweeps: 1) denial 2) anger 3) bargaining 4) depression 5) aggressively re-reading the acquisition faq section on severance packages.
-
ruchowdh.bsky.social reposted
Responsible Artificial Intelligence LLC
@llc_rai
Public comments on the proposed Rules for the New York City law on automated employment decision tools (AEDT) have been posted
☛ View our comments here:
rules.cityofnewyork.us/rule/automated-employment-decision… -
ruchowdh.bsky.social reposted
BiasBounty
@BiasBounty
Ahoy! Here be treasure!
Send us yr most dastardly, swashbuckungly gnarly pirate pictures and win th’ bounty!
Kindles! Nintendos! Swag! All could be yers, and no artist be turned away!
We judgin’ on yer creativity more than yer artistic abilities. Submit until Friday! -
ruchowdh.bsky.social
@ruchowdh
Our first fun challenge! We’ll send you a kindle or an 8bit Nintendo if you win 📕
BiasBounty
@BiasBounty
Ahoy! Here be treasure!
Send us yr most dastardly, swashbuckungly gnarly pirate pictures and win th’ bounty!
Kindles! Nintendos! Swag! All could be yers, and no artist be turned away!
We judgin’ on yer creativity more than yer artistic abilities. Submit until Friday! -
ruchowdh.bsky.social reposted
wells (wants a better world)
@WellsLucasSanto
Gonna try something out... if you're moving from Twitter to Mastodon (or at the very least, have started to use an account there), post on Mastodon using the #twittermigration hashtag to help folks find you!
-
ruchowdh.bsky.social reposted
Scott Galloway
@profgalloway
When you say you’ll lay off 75% of your staff, the first to leave are the 25% you want to keep, because they have the most options #elon #twitter
-
ruchowdh.bsky.social reposted
NYTPitchbot
@DougJBalloon
How Tesla became the Axe body spray of cars
-
ruchowdh.bsky.social
@ruchowdh
Ok so after an initial few days of feeling strange I opened @joinmastodon and it felt friendlier. It’s kind of nice to re-curate your feed. Come find me and the others!
mastodon.social/@ruchowdh -
ruchowdh.bsky.social reposted
The Tech Humanist
@thetechhumanist
Aside from being interesting coverage of the "bias bounty" approach to removing prejudice from AI, this article quotes two of our previous #thetechhumanistshow guests: @ruchowdh and @atg_abhishek. 👏 👏
www.technologyreview.com/2022/10/20/1061977/ai-bias-bount… -
ruchowdh.bsky.social
@ruchowdh
100? My sweet summer child this is me now
-
ruchowdh.bsky.social reposted
Avijit Ghosh
@evijit
This Thursday (27th Oct) 12:30pm ET I am going to give a lightning talk on "Subverting Fair Image Search with Generative Adversarial Perturbations" as a part of the 'Celebrating Young Researchers' event at @trustworthy_ml's 2-year Anniversary Symposium!
us02web.zoom.us/webinar/register/WN_a5anHmDaRIWY6xYupiD0mg -
ruchowdh.bsky.social reposted
Evan Meagher
@evanm
“I could run Twitter with 25% of the staff” is the new “I could build Twitter in a weekend”
-
ruchowdh.bsky.social reposted
Dr. Sarah T. Roberts ubiquity75@mastodon.social
@ubiquity75
Ubiquity75@mastodon.social
-
ruchowdh.bsky.social reposted
Kristian Lum
@KLdivergence
Feels like a good day to rebuild from scratch.
I'm KLdivergence@mastodon.social Mastodon.
Find me there, and I'll follow back. -
ruchowdh.bsky.social
@ruchowdh
I’m sad to be taking this pinned tweet down. Frankly I forgot about it but every time I get a new wave of followers this seems to resonate with a handful of them.
ruchowdh.bsky.social
@ruchowdh
By me and @gleemie on tech’s quiet colonization of the social sciences:
“The beautiful messiness of humanity cannot be quantified into data columns, programmed, scaled, and engineered.”
www.wired.com/story/tech-needs-to-listen-to-actual-resear… -
ruchowdh.bsky.social
@ruchowdh
Slowly making the migration to @joinmastodon
Find me under the same handle mastodon.social/@ruchowdh -
ruchowdh.bsky.social reposted
New_ Public | BSKY: @newpublic.substack.com
@WeAreNew_Public
A unique approach - a bounty for identifying bias in AI algorithms - it takes a digital village to build equitable digital spaces!
ruchowdh.bsky.social
@ruchowdh
The 8 bit bias bounty is now live!! Thank you @Melissahei for the article on what the bounty program means in context of the evolving AI regulatory space and for community based feedback - www.technologyreview.com/2022/10/20/1061977/ai-bias-bount…
-
ruchowdh.bsky.social
@ruchowdh
If you’re a manager at Twitter, now’s the time to reflect on your values as an individual and determine who it is you’re here for.
-
ruchowdh.bsky.social reposted
Subho Majumdar | শুভব্রত মজুমদার
@sbmisi
We're live!! Sign up for @biasbounty today to be part of a new way of doing responsible ML through practitioner enablement.
Thanks @Melissahei for the article, and I'm glad to see @sarahookr and @atg_abhishek make appearances too!
ruchowdh.bsky.social
@ruchowdh
The 8 bit bias bounty is now live!! Thank you @Melissahei for the article on what the bounty program means in context of the evolving AI regulatory space and for community based feedback - www.technologyreview.com/2022/10/20/1061977/ai-bias-bount…
-
ruchowdh.bsky.social reposted
Camlis Org
@CamlisOrg
Next session kicks off with this announcement and more details!
ruchowdh.bsky.social
@ruchowdh
We are launching officially today @CamlisOrg so pls find me or @sbmisi if you have any questions!
If you’d like to host your own bounty, sponsor or volunteer, please reach out! -
ruchowdh.bsky.social reposted
Charlotte Jee
@charlottejee
Cool piece by @Melissahei delving into a new AI bias bounty competition launched today by @ruchowdh et al.
Expect to see more of this sort of thing as AI audits become more commonplace.
www.technologyreview.com/2022/10/20/1061977/ai-bias-bount… -
ruchowdh.bsky.social
@ruchowdh
The 8 bit bias bounty is now live!! Thank you @Melissahei for the article on what the bounty program means in context of the evolving AI regulatory space and for community based feedback - www.technologyreview.com/2022/10/20/1061977/ai-bias-bount…
ruchowdh.bsky.social
@ruchowdh
We address algorithmic bias at its source - the data - biasbounty.ai/8-bbb
The task: create an image tagging model that minimizes demographic biases. We have $16,000 in prizes and fun swag (like kindles and pirate rubber duckies!)
ruchowdh.bsky.social
@ruchowdh
We are launching officially today @CamlisOrg so pls find me or @sbmisi if you have any questions!
If you’d like to host your own bounty, sponsor or volunteer, please reach out! -
ruchowdh.bsky.social
@ruchowdh
So my brief time was plagued with connectivity issues so I wanted to share what I was (trying to) share on this NIST panel!
In short, there’s adventurous new ways to explore participatory design - sections 5.1 and 5.2 of the @NIST RMF framework 1/Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
We aim to improve community driven engagement in algorithmic accountability. Aligned with NISTs RMF framework, these initiatives aim to enable focused, actionable, and inclusive feedback mechanisms that enable algorithmic transparency and accountability.
-
ruchowdh.bsky.social reposted
Subho Majumdar | শুভব্রত মজুমদার
@sbmisi
For those who can't be in @CamlisOrg in-person, you can tune in online to watch @ruchowdh and I launch the first ever third-party Algorithmic @BiasBounty!
Hyrum Anderson
@drhyrum
Can’t travel to @CamlisOrg for the in-person event and reunion with professional colleagues? We’ve got you covered.
Join the livestream free from anywhere in the world for the most relevant content at the intersection of ML+Security.
sandscapital.zoom.us/webinar/register/WN_VJundFhXSuK-X_xU… -
ruchowdh.bsky.social reposted
ruchowdh.bsky.social
@ruchowdh
It’s almost here! @sbmisi and I will be at @CamlisOrg to launch Bias Buccaneers first bias bounty, an ambitious challenge to address data labeling biases.
Sign up below ⬇️
BiasBounty
@BiasBounty
Arrrrrrrrrr ye prepared for thee first mutiny? This be th’ week we begin t’ rid thee world o’ algorithmic bias.
You and yer crew should join today…or else!
biasbounty.ai/the-code -
ruchowdh.bsky.social
@ruchowdh
It’s almost here! @sbmisi and I will be at @CamlisOrg to launch Bias Buccaneers first bias bounty, an ambitious challenge to address data labeling biases.
Sign up below ⬇️
BiasBounty
@BiasBounty
Arrrrrrrrrr ye prepared for thee first mutiny? This be th’ week we begin t’ rid thee world o’ algorithmic bias.
You and yer crew should join today…or else!
biasbounty.ai/the-code -
ruchowdh.bsky.social reposted
Dr. Casey Fiesler is no longer on here
@cfiesler
PhD applicants (and friends of applicants) in computing-adjacent fields: this is a thread of tweets about open positions/faculty recruiting/programs. Probably mostly InfoSci/HCI/responsible tech/social computing. Please share, and also give me more to add! 🙏
-
ruchowdh.bsky.social reposted
BiasBounty
@BiasBounty
Arrrrrrrrrr ye prepared for thee first mutiny? This be th’ week we begin t’ rid thee world o’ algorithmic bias.
You and yer crew should join today…or else!
biasbounty.ai/the-code -
ruchowdh.bsky.social reposted
Luca (also on the other platforms)
@__lucab
senior* ic folks i have some questions for you!
how do you balance your IC work vs mentoriong / helping others? what's the share of each?
*defined as someone who's supposed to have indriect impact—regardless of actual title -
ruchowdh.bsky.social reposted
Kristian Lum
@KLdivergence
Woke up to some messages this morning that I (my picture) was on NOVA! And it looks like @wsisaac makes an even bigger appearance with a talking role! How cool! Scrolling through, I also saw @Sarah_Brayne & @ruha9 . Will def be watching tonight.
www.pbs.org/video/computers-v-crime-um7cco/ -
ruchowdh.bsky.social reposted
PCLOB
@PCLOB_GOV
PCLOB is hiring! If you're interested in applying for the position of Executive Director, click here to learn more: www.usajobs.gov/job/681610200
-
ruchowdh.bsky.social
@ruchowdh
I’d almost rather a dystopian apocalypse than whatever the fuck this is
Tom Warren
@tomwarren
Microsoft is partnering with Meta in to bring Teams, Windows, Office, and even Xbox Cloud Gaming to Quest VR headsets. It's the biggest Microsoft and Meta partnership since the Facebook and Windows Phone integration a decade ago. Full details here: www.theverge.com/2022/10/11/23397251/meta-microsoft-partn…
-
ruchowdh.bsky.social reposted
AI Vulnerability Database
@AvidMldb
Join us at 7pm EST for our weekly Brainstorm! Tonight we will be discussing our collaboration with @IncidentsDB to see how our projects can work together to make the world a better place. #artificialintelligence #aiid #avid #aiethics #cybersecurity #techforgood #ai #brainstorm
-
ruchowdh.bsky.social reposted
Will Oremus
@WillOremus
“Content moderation” was once industry jargon. Now it’s a front line in the culture war and the subject of multiple potential Supreme Court cases.
Here’s my attempt to trace the story of how that happened, from 1996 to today, in 3,000 words. www.washingtonpost.com/technology/2022/10/09/social-media… -
ruchowdh.bsky.social reposted
Samanth Subramanian
@samanth_s
For ~a year, I've dug into the Hindu right's impact on Bollywood, to find out how the vauntedly liberal industry - directors, platforms, writers, actors - is responding.
Nearly 50 interviews later, here's the piece. This is not a pretty picture... [1]
www.newyorker.com/magazine/2022/10/17/when-the-hindu-righ… -
ruchowdh.bsky.social reposted
Natalia Domagala
@NaDomagala
#algorithmictransparency news - I'm thrilled to see how my fantastic ex-colleagues at the CDDO and @CDEIUK have been advancing the work on the UK Algorithmic Transparency Standard. More in this blog, summary below 🧵 cdei.blog.gov.uk/2022/10/10/developing-the-algorithmic-tr…
-
ruchowdh.bsky.social reposted
Tim Howles
@AimeTim
1. It is with great sadness that we learn the news of Bruno Latour’s passing today. He was an intellectual giant, a friend, & someone whose work inspired me greatly. A thread for those who want to know more about this extraordinary life.
-
ruchowdh.bsky.social
@ruchowdh
No context Texas
-
ruchowdh.bsky.social reposted
ACM EAAMO (Handle Retired, Visit @EAAMO_ORG)
@ACMEAAMO
The final paper of the session is Participation Is not a Design Fix for Machine Learning by @mona_sloane, @mannymoss, @awomoloolaitan, and @laura4lano! The talk is presented virtually over Hopin 🧵(1/4)
-
ruchowdh.bsky.social reposted
ladies and gentlemen, the weekend 😌
@CraigWeekend
-
ruchowdh.bsky.social reposted
Responsible Artificial Intelligence LLC
@llc_rai
Announcing This Month in Responsible AI, a newsletter summarizing developments in research, policy and industry. This month: the formation of RAI LLC, NYC's Proposed Rules for the law on automated employment decision tools (AEDT), and @BiasBounty raillc.substack.com/p/2022-10-announcing-rai-llc
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
I'm thrilled to announce the launch of @llc_rai, my new consultancy focused on practical, yet responsible AI solutions for the evolving needs of businesses and regulators that is informed by the very latest academic research.
-
ruchowdh.bsky.social reposted
Shira Ovide
@ShiraOvide
I mean, props to the Twitter employee who referenced Dunning-Kruger, via @lizzadwoskin & @WillOremus.
Not your normal blind quote.
www.washingtonpost.com/technology/2022/10/05/twitter-elon… -
ruchowdh.bsky.social
@ruchowdh
If Twitter is Pawnee town hall then Twitter META is the parks department ht @runefeather
-
ruchowdh.bsky.social reposted
parker lyons
@tweetsbyparker
getting through my meetings today
x.com/DukeSilver00/status/1420911079522193417/video/1
unusual_whales
@unusual_whales
BREAKING
Tesla CEO Musk said to propose to Twitter, $TWTR, to proceed with deal at $54.20 -
ruchowdh.bsky.social
@ruchowdh
Living the plot of succession is fucking exhausting
ruchowdh.bsky.social
@ruchowdh
I am sitting on 2023 company wide strategy readouts and I guess we are going to collectively ignore what’s going on
ruchowdh.bsky.social
@ruchowdh
This in real time
ruchowdh.bsky.social
@ruchowdh
This is going to sound funny but Musks immediate chilling effect was something that bothered me significantly. Twitter has a beautiful culture of hilarious constructive criticism, and I saw that go silent because of his minions attacking employees.
-
ruchowdh.bsky.social reposted
Alondra Nelson
@AlondraNelson46
Join us this morning at 10am ET at wh.gov for a discussion of the Blueprint for an AI Bill of Rights and the announcement of key Biden-Harris Administration actions to ensure technology is serving and benefiting the American people. www.whitehouse.gov/ostp/news-updates/2022/10/04/blueprint…
-
ruchowdh.bsky.social reposted
Veena Dubal
@veenadubal
This is an amazing long form piece on caste discrimination in the universities educating Indian engineers funnel into US-based Big Tech caste discrimination. www.bloomberg.com/news/features/2021-03-11/how-big-tech-i…
-
ruchowdh.bsky.social reposted
AISTATS Conference
@aistats_conf
The AISTATS 2023 Call for Papers is out!
aistats.org/aistats2023/call-for-papers.html
Submission Deadline: Thu Oct 13, AoE
Program Chairs: Jennifer Dy, Jan-Willem van de Meent (@jwvdm)
General Chair: Francisco Ruiz
Workflow Chairs: Christian Naesseth (@chris_naesseth), Davin Hill (@davinhill_) -
ruchowdh.bsky.social reposted
NYU Tandon
@nyutandon
#tandon professor @mona_sloane's #thisisnotadrill exhibition debuted at Mamdouha Bobst Gallery with great success on 9/28. It combines tech, equity and the climate emergency and will be open for viewing until 12/4/22. #nyutandonmade @tisch @NYULibraries
tisch.nyu.edu/fif/current-projects/this-is-not-a-drill -
ruchowdh.bsky.social reposted
AI Vulnerability Database
@AvidMldb
Hello World! AVID has just started our latest community feature our Blog.
www.avidml.org/blog/hello-world
Keep a watch out for upcoming entries and join us to create more.
#ai #artificialintelligence #vulnerability #security #helloworld #blog #avid -
ruchowdh.bsky.social reposted
Subho Majumdar | শুভব্রত মজুমদার
@sbmisi
The first @AvidMldb blog post is up. More exciting stuff coming very soon!!
AI Vulnerability Database
@AvidMldb
Hello World! AVID has just started our latest community feature our Blog.
www.avidml.org/blog/hello-world
Keep a watch out for upcoming entries and join us to create more.
#ai #artificialintelligence #vulnerability #security #helloworld #blog #avid -
ruchowdh.bsky.social reposted
Nathan Baschez
@nbaschez
-
ruchowdh.bsky.social
@ruchowdh
If you are in NYC make the time to go see this exhibit!
-
ruchowdh.bsky.social
@ruchowdh
Reposting this actual life goal achieved. Proud of Twitter METAs collaboration with Openmined!
OpenMined
@openminedorg
We're very excited to work w/ New Zealand & US Govs, @Microsoft, & @Twitter on expanding #pysyft use helping independent researchers study major #ai algorithms
👇New Zealand PM @jacindaardern & French President @EmmanuelMacron announce the project
WATCH: www.youtube.com/watch?v=tZ0u-Bt83mE -
ruchowdh.bsky.social reposted
BiasBounty
@BiasBounty
Arrrrr. It’s Monday. The shared scourge o’ landlubbers and pirates alike.
It be a good day to sign up fer our first mutiny at biasbounty.ai. -
ruchowdh.bsky.social
@ruchowdh
I love this - is this a standard part of education? Are there options available for younger kids? 9th grade is years after kids have developed their internet habits.
-
ruchowdh.bsky.social reposted
𝙽𝚒𝚌𝚔 𝙼𝚊𝚝𝚑𝚎𝚜𝚘𝚗
@willplayforfood
I want to take a minute to acknowledge @williams_jutta and @iamtrask for the long ago work to get this going.
And @ruchowdh, the Twitter META team and the many new faces/partners for all the work to bring it to where it is today.
www.miragenews.com/christchurch-call-initiative-on-algori… -
ruchowdh.bsky.social reposted
Luca (also on the other platforms)
@__lucab
reminder that this is happening tomorrow!
see you there?
Fair Recommendation
@FAccTRec
The #recsys2022 FAccTRec workshop will be held in Tomorrow!
It's a venue to discuss responsible operation of RS. So many exciting invited and contributed talks are waiting for you.
We look forward your participation!
facctrec.github.io/facctrec2022/program/
@facctrec -
ruchowdh.bsky.social
@ruchowdh
This week at the UNGA, PM Ardern announced the Christchurch Call Initiative on Algorithmic Outcomes - an expansion of Twitter META's PETs work with @openminedorg that now includes Microsoft, the US and NZ govts.
Global Government Affairs
@GlobalAffairs
As the online violent extremism landscape continues to evolve, the #christchurchcall remains as critical as ever.
We must continue to strengthen our efforts to address these challenges in a way that protects a free, Open Internet.
ruchowdh.bsky.social
@ruchowdh
Our hope is that our good faith effort to improve access while also respecting data privacy will set the tone for further partnerships.
Global Government Affairs
@GlobalAffairs
Our ML Ethics, Transparency, and Accountability (META) team will drive this work, in collaboration with @openminedorg.
In Phase 1, our goal is to allow access to the dataset underpinning our algorithmic amplification paper via privacy enhancing technologies. -
ruchowdh.bsky.social reposted
Subho Majumdar | শুভব্রত মজুমদার
@sbmisi
I'm pleased to announce the AI Vulnerability Database @AvidMldb, a grassroots, open-source effort to build a common standard for @trustworthy_ml practitioners to codify and track ML failures.
1/ -
ruchowdh.bsky.social reposted
NYU Alliance for Public Interest Technology
@NYU_Alliance
AI, Law & Ethics: Learn how to comply with emerging regulations, avoid legal risks, and protect your company and ethics. @Lawgeek #executiveeducation #nyualliance
-
ruchowdh.bsky.social reposted
Kristian Lum
@KLdivergence
I was named one of @RiceUniversity's Outstanding Young Engineering Alumni. This is exciting for 3 reasons. (1) It feels wonderful to be acknowledged by my alma mater, (2) they still think I'm "young"!! 😂, and (3) I get to meet the host, @Dr_TalithiaW!
engineering.rice.edu/news/rice-engineering-honor-alumni-a… -
ruchowdh.bsky.social reposted
Olivier Schulbaum
@SchulbaumO
Mutiny in heaven is my favourite Bad seeds song ever. Mutiny in AI is the next move to take! Rise up against algorithmic bias. #ai #ethics
ruchowdh.bsky.social
@ruchowdh
Last year, @williams_jutta and I pioneered the Algorithmic Bias Bounty as Twitter META. We hoped our idea would spread - and it did. What's next is a third space to do this work, and with @sbmisi @ScottFromNY and Ben Colman, we're excited to announce
www.biasbounty.ai -
ruchowdh.bsky.social reposted
The Information Society Project
@yaleisp
We're starting our Fall 2022 Law & Technology speaker series with a bang! Don't miss @sushmaraman on "The Coming Good Society: Why New Realities Demand New Rights"
Tuesday, September 13 at 12pm ET
Sterling Law Building 128
cc @CarrCenter @YaleLawSch -
ruchowdh.bsky.social
@ruchowdh
Last year, @williams_jutta and I pioneered the Algorithmic Bias Bounty as Twitter META. We hoped our idea would spread - and it did. What's next is a third space to do this work, and with @sbmisi @ScottFromNY and Ben Colman, we're excited to announce
www.biasbounty.ai
ruchowdh.bsky.social
@ruchowdh
Our new project is a non-profit entity aimed to engage the community with corny pirate jokes and meaningful progress in the field of algorithmic auditing. We'll be announcing our first challenge at @CamlisOrg, where we will be testing frameworks developed by @AvidMldb
ruchowdh.bsky.social
@ruchowdh
This is a passion project, independent of our work with our respective employers, driven by our motivation to cultivate a community of practice. We'll be dropping more details over the next few weeks, so stay tuned!
BiasBounty
@BiasBounty
Ahoy!
Set sail with us next month when we begin our maiden voyage. 🚢 Anchors away at CAMLIS on 10/20/22 ⚓️
Our first Bias Bounty. Our first Mutiny! 🏴☠️
Sign up to be a Bias Buccaneer at: biasbounty.ai -
ruchowdh.bsky.social reposted
Special Competitive Studies Project
@scsp_ai
The #scsptech Report Mid-Decade Challenges to National Competitiveness is live! Thank you to all for months of tireless work for what we believe is a product that will ensure that the U.S. is positioned to win the tech competition between now & 2030! www.scsp.ai
-
ruchowdh.bsky.social reposted
David Robinson
@dgrobinson
Before there were many high stakes algorithms – before “AI ethics” was a thing – a doctor named Belding Scribner became an unwitting AI ethics pioneer. His story is out now in Slate, part of an excerpt from my new book. slate.com/technology/2022/08/kidney-allocation-algorithm-…
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
I am no longer involved with Parity Technologies, Inc. in any capacity. Stay tuned for what's next!
-
ruchowdh.bsky.social reposted
-
ruchowdh.bsky.social reposted
ScienceInsider
@ScienceInsider
BREAKING: White House issues new policy that will require, by 2026, all federally-funded research results to be freely available to public without delay, ending longstanding ability of journals to paywall results for up to 1 year. Coverage coming on @ScienceInsider.
-
ruchowdh.bsky.social
@ruchowdh
Congratulations @zywind
Vera Liao
@QVeraLiao
📢 Happy to share this preprint, now accepted at #hcomp2022. We introduce a perspective of contextualized evaluation for explainable AI, hoping to close the gaps between algorithmic research and user requirements in diverse downstream usage contexts
arxiv.org/abs/2206.10847
🧵1/4 -
ruchowdh.bsky.social
@ruchowdh
It had been some weeks since I woke up to my phone piling in the text messages; knew it couldn’t last forever.
Never a dull day at the bird factory 🐦 -
ruchowdh.bsky.social reposted
Ian Shapiro
@IonShapiro
Anyone teaching research methods with R. Highly recommend this new edition that uses tidyverse. User friendly, lots of student resources.
-
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
I hate when a post of mine goes this big so going to mute. For the people who disagree w me:
ruchowdh.bsky.social
@ruchowdh
Mm so I listened and I still do think the above. Here’s why - there’s this inevitability built into his language that I think is incorrect. He also still refers to “the people” meaning not himself, even though he has no clue about these systems being built.
ruchowdh.bsky.social
@ruchowdh
Mm so I listened and I still do think the above. Here’s why - there’s this inevitability built into his language that I think is incorrect. He also still refers to “the people” meaning not himself, even though he has no clue about these systems being built.
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
A BIG step forward for LGBTQ+ rights in Singapore. But there's so much more that needs to happen for non-straight people to live lives that vaguely resemble normalcy. Acceptance of same sex families needs to reflect in updates to housing policy, healthcare policy...
-
ruchowdh.bsky.social
@ruchowdh
Trying to figure out how to send memes efficiently across multiple unrelated signal groups and realizing that’s what social media is
-
ruchowdh.bsky.social reposted
Lina Khan
@linamkhan
1. Firms track & collect data on Americans at a stunning scale—on our location, our health, what we read online, who we know, what we buy.
@FTC is seeking comment on whether to issue rules aimed at commercial surveillance & lax data security practices.
www.ftc.gov/news-events/news/press-releases/2022/08/ftc-e… -
ruchowdh.bsky.social
@ruchowdh
Recalling the time this woman burst into tears defending Elon’s genius because I told her he was the worlds smartest grifter.
Oh, her husband had paid to fly spaceX as a passenger 🙄 -
ruchowdh.bsky.social
@ruchowdh
Cast your vote!
Avijit Ghosh
@evijit
@anxiousglxtch and I have applied to speak at #sxsw2023 and we need your vote!
Our talk is (provocatively) titled: "GAN “Art” Devalues Human Artistry - the HeLa cell debate in the AI Art Space". A thread: 🧵
panelpicker.sxsw.com/vote/122773 -
ruchowdh.bsky.social reposted
Avijit Ghosh
@evijit
@anxiousglxtch and I have applied to speak at #sxsw2023 and we need your vote!
Our talk is (provocatively) titled: "GAN “Art” Devalues Human Artistry - the HeLa cell debate in the AI Art Space". A thread: 🧵
panelpicker.sxsw.com/vote/122773 -
ruchowdh.bsky.social reposted
Engineering
@Engineering
Sharing a big congrats to our very own @KLdivergence who received the Committee of Presidents of Statistical Societies Leadership Academy award for emerging leaders in statistics at today’s #jsm2022 conference! 🥳
-
ruchowdh.bsky.social reposted
David Robinson
@dgrobinson
I'd like to tell you a story about how people built a life-or-death algorithm in a relatively inclusive, accountable way. It shows the promise and limits of our efforts to democratize AI. It's at the heart of my new book, Voices in the Code, out Sept. 8.
www.amazon.com/Voices-Code-People-Values-Algorithm-dp-087… -
ruchowdh.bsky.social reposted
Luca (also on the other platforms)
@__lucab
a few weeks back i partecipated in an informal gathering / workshop among folks doing responsible ai in the field
thank you @baxterkb for organizing and sharing the final perspectives and thanks to @scuethics for hosting us:
blog.salesforceairesearch.com/responsible-ai-matters-how-… -
ruchowdh.bsky.social
@ruchowdh
Celebrating @KLdivergence s distinguished career award for her work advancing the field of ML ethics. It’s an honor to have you with Twitter META 🎉
magazine.amstat.org/blog/2022/05/02/copss-leadership-acad… -
ruchowdh.bsky.social reposted
moo
@moo_hax
We have ~21 challenges donated by ~5 orgs that span all manner of ML attacks. Big shoutout to @GTKlondike @josephtlucas @JankhJankh @rharang @comathematician @ColdwaterQ @BenevOrang for their time and effort.
Stay tuned for more details!
AI Village @ DEF CON
@aivillage_dc
The AI Village is proud to announce our CTF for @defcon 30. Evade, poison, steal, and otherwise hack AI and ML systems. There'll be something for everybody, so check it out. Follow for more details.
-
ruchowdh.bsky.social
@ruchowdh
Whenever I see articles like this I think about affordances. There is a lot of discussion about access and resources, but not enough about failures. Who is allowed to fail? What are the consequences? When is failure a “learning moment” vs a condemnation?
blockworks.co/meta-metaverse-unit-has-lost-5-7b-this-year/ -
ruchowdh.bsky.social
@ruchowdh
Who’s going to be at black hat?
-
ruchowdh.bsky.social reposted
Scott Hechinger
@ScottHech
This is astonishing, yet not surprising. If it feels like you're far less safe in NYC than ever before, it's not bc shootings are up. It's because the media (fueled by lies & fearmongering by NYC Mayor Eric Adams) is up. Look at this chart. Red line is shootings. Grey is media.
-
ruchowdh.bsky.social
@ruchowdh
“Prostitution” was shockingly not on my “things tech bros reinvent” list
-
ruchowdh.bsky.social reposted
No Lie with Brian Tyler Cohen
@NoLieWithBTC
This new commercial from Mothers Against Greg Abbott is one of the best political ads I’ve ever seen.
This media couldn't be displayed. See it on the original post. -
ruchowdh.bsky.social reposted
aimee
@kingosunens
a thread of lgbtq+ bollywood films because every time i see one of these threads, i never see us on there :/ hoping this will make the gay desi community feel seen + introduce non desis to new cinema!
-
ruchowdh.bsky.social
@ruchowdh
This is very very bad.
Corey Quinn
@QuinnyPig
"Why is @Amazon acquiring @OneMedical bad news?"
Let's explore.
1. Horizontal integration at Amazon scale of a company that now has the right to access my medical records under HIPAA opens up a huge problem for abuse. -
ruchowdh.bsky.social reposted
Avijit Ghosh
@evijit
Little known fact but people who are/have worked with the Twitter META team are all hot af 🔥
Liz Bright Marquis
@lizbmarquis
both looking handsome!! so sad i'm not there
-
ruchowdh.bsky.social
@ruchowdh
Oo I want to host a dall-e bias bounty! Who’s in?
-
ruchowdh.bsky.social
@ruchowdh
Dr. Pepper
Dr. Lechter
Dr. Cat
Dr. Frankenstein
Dr. Nick -
ruchowdh.bsky.social reposted
Ethan Mollick
@emollick
Of all of the “dangers of AI” papers, this is most worrying: AI researchers building a tool to find new drugs to save lives realized it could do the opposite, generating new chemical warfare agents. Within 6 hours it invented deadly VX… and worse things www.nature.com/articles/s42256-022-00465-9
-
ruchowdh.bsky.social
@ruchowdh
“This is nothing like polling, we are trying to get a number of representative accounts so we can understand what’s happening across the whole population. NOTHING LIKE POLLING!!!”
Hello and welcome to statistics, in todays lesson we discuss sampling -
ruchowdh.bsky.social
@ruchowdh
Top 10 tips for avoiding top 10 tips👇
-
ruchowdh.bsky.social reposted
Privacy Matters
@PrivacyMatters
The poverty of privacy. Supermarket loyalty cards.
For many families, these savings are crucial. -
ruchowdh.bsky.social reposted
GLAAD
@glaad
GLAAD President and CEO @sarahkateellis spoke to @Forbes this week about the new Social Media Safety Index report and the harms of anti-LGBTQ rhetoric on the platforms. www.forbes.com/sites/dawnstaceyennis/2022/07/13/report-th…
-
ruchowdh.bsky.social reposted
Karen Hao
@_KarenHao
The latest mind-bending details of one of history's largest data heists: Experts say the Shanghai police database, which exposed nearly 1b Chinese citizens' info, didn't just lack a password—it was using such outdated software, no password could be added. www.wsj.com/articles/alibaba-executives-called-in-by-chin…
-
ruchowdh.bsky.social reposted
Mrs Whatsit's burner phone
@MadeleineLEngle
“I didn’t mean to tell you,” Mrs Whatsit faltered. “I didn’t mean ever to let you know. But, oh, my dears, I did so love being a star!”
#webb
A Wrinkle in Time -
ruchowdh.bsky.social reposted
Casey Newton
@CaseyNewton
One of the better exhibits I’ve seen in a legal filing
-
ruchowdh.bsky.social
@ruchowdh
Ruh roh looks like the armchair Twitter CEOs are emerging again
-
ruchowdh.bsky.social
@ruchowdh
Excited to see Twitter META's bias bounty work carrying forward. Looking forward to participating in the advisory board for this challenge, and I hope to see more in the future!
Stanford HAI
@StanfordHAI
Introducing the #aiauditchallenge – a $71K competition to design better AI audits. @StanfordHAI & @StanfordCyber are seeking proposals for tools that enable regulators, journalists & civil society to effectively audit deployed AI systems/open source models.hai.stanford.edu/ai-audit-challenge
-
ruchowdh.bsky.social reposted
joyojeet pal
@joyopal
For students considering a PhD in the HCI or associated fields, if you don't see being a professor/academic when you finish, seriously reconsider doing the PhD.
At the very least, definitely reconsider mentioning this straight off to someone who you may want as an advisor. -
ruchowdh.bsky.social
@ruchowdh
The best part about Twitter is that I have a forum for my l'esprit de l'escalier, which is a fancy way of saying “things I should know better than to say out loud”
-
ruchowdh.bsky.social
@ruchowdh
How do we get some of this energy in the US
-
ruchowdh.bsky.social
@ruchowdh
Ding dong, as they say
Alex Weprin
@alexweprin
Flash: Elon Musk tells Twitter that he is terminating the proposed acquisition: www.sec.gov/Archives/edgar/data/1418091/00011046592207841…
-
ruchowdh.bsky.social reposted
Alex Weprin
@alexweprin
Flash: Elon Musk tells Twitter that he is terminating the proposed acquisition: www.sec.gov/Archives/edgar/data/1418091/00011046592207841…
-
ruchowdh.bsky.social
@ruchowdh
I hope that all of us can live life like this man, and frankly humanity would be better off for it.
However I can’t help but get strong Walden Pond vibes here. A life like this doesn’t come without someone doing a lot of labor to support it.
www.quantamagazine.org/june-huh-high-school-dropout-wins-… -
ruchowdh.bsky.social reposted
Vincent Manancourt
@vmanancourt
BREAKING: Europe facing Facebook blackout as Ireland orders halt to its EU-US data transfers.
FB has repeatedly said such an order could force it to shutter services in EU. Other EU regulators now have 1 month to weigh in before decision finalized.
pro.politico.eu/news/151897 -
ruchowdh.bsky.social reposted
Mark Bergen
@mhbergen
Twitter sues Indian government. YouTube caves to BJP and removes a video titled "Anthem for Kashmir." www.medianama.com/2022/07/223-youtube-blocks-anthem-for-k…
-
ruchowdh.bsky.social reposted
Ferenc Huszár
@fhuszar
If you live in the uk and are eligible to be a host under the Homes for Ukraine scheme, please reach out. We need hosts for a couple of talented kids and their families (we actually have funding to move them to rented accommodation once in the UK so they only need a landing pad)
-
ruchowdh.bsky.social
@ruchowdh
Given that the last time I test tried one it thought I was 18-22 (I am twice that age)…yes.
HALT AI
@HALT_AI
As Instagram tests new ways to verify users' ages, could #aibias be an issue they’ll have to contend with? edition.cnn.com/2022/06/27/tech/instagram-ai-age-estimati…
-
ruchowdh.bsky.social reposted
Luca Bertuzzi
@BertuzLuca
BREAKING Both the #dma and #dsa have been adopted in the European Parliament with a very broad majority.
-
ruchowdh.bsky.social
@ruchowdh
Grateful for the (minor) revisions made thanks to @emilymbender s popular thread but baffled at why an intern was given this piece. Note that the original headline and article didn’t say “claims”.
Where were the editors on this?
www.bloomberg.com/news/articles/2022-06-30/new-algorithm-… -
ruchowdh.bsky.social reposted
aaron is vibe shifting to mastodon
@binary_aaron
web3 should be renamed to web𝒊 because it's imaginary
-
ruchowdh.bsky.social reposted
Bayt Al Fann
@BaytAlFann
Eid al-Adha is celebrated on the third day of Hajj & lasts for 3 days. Next weekend Muslims across the world will celebrate Eid al-Adha with a host of traditional sweet treats
Here are 24 sweet delicacies different Muslim cultures make for this special time of year…
A thread… -
ruchowdh.bsky.social reposted
The Tech Humanist
@thetechhumanist
Our latest episode was delayed by the (all-too-)human experience of fighting off covid but we're back up & running:
"Why Human Experience?"
featuring @CharlieCole @NeilRedding @CathyHackl @aexm @ruchowdh @MarcusWhitney @techethicist #thetechhumanistshow
www.thetechhumanist.com/2022/07/01/why-human-experience-v… -
ruchowdh.bsky.social reposted
ACM FAccT
@FAccTConference
Ok #facct22 attendees we want to hear from you! Fill out our survey and help us figure out what worked and what didn't this FAccT. This *will* help shape FAccT's future—and you get a chance of winning some swag—so get on it! 🚀🚀🚀🧐🧐🧐📝📝📝einz4o183fn.typeform.com/facct22
-
ruchowdh.bsky.social
@ruchowdh
Correct - "The basic assumption when using a machine-learning model is that there is a pattern to be identified; in this case, that there’s some “normal” behavior that shooters exhibit at the scene of the crime. But finding such a pattern is unlikely."
www.wired.com/story/school-surveillance-never-protect-kid… -
ruchowdh.bsky.social reposted
Meg Young
@megyoung0
I am very excited to be joining people I admire @datasociety as a Participatory Methods Research Fellow
I'll be on the AI on the Ground Team who do phenomenal work to inform policy---including calling for participation in impact assessments & commissioning Global Majority work -
ruchowdh.bsky.social reposted
Karen Hao
@_KarenHao
I wrote about a topic I’ve been itching to address for some time: how AI PR hype, coupled with increasingly flashy AI-generated outputs, is creating a dangerous gap between the technology’s perception & reality. A collab with @MilesKruppa. www.wsj.com/articles/tech-giants-pour-billions-into-ai-bu…
-
ruchowdh.bsky.social
@ruchowdh
There’s now been three METAminis born on my team this year, and it always amazes me how much each birth brings feelings of hope, renewed mission, and desire to do good in a world that seems rather dark at the moment.
-
ruchowdh.bsky.social
@ruchowdh
This is…a bias bounty…right?
Ethan Perez
@EthanJPerez
We’re announcing the Inverse Scaling Prize: a $100k grand prize + $150k in additional prizes for finding an important task where larger language models do *worse*.
Link to contest details: github.com/inverse-scaling/prize
🧵
ruchowdh.bsky.social
@ruchowdh
Glad to see it happening and at scale - I just hope that the effort is channeled into supporting positive change where it’s helpful, not just where it’s profitable
-
ruchowdh.bsky.social reposted
Sara K. Runnels
@omgskr
My ideal summer body is one that I have complete control over
-
ruchowdh.bsky.social reposted
aaron is vibe shifting to mastodon
@binary_aaron
content warning: sincerity.
I've been doing fairness or t&s style work off and on in industry for the last 6~ years or so and haven't really participated much in the academic side of it until very recently. -
ruchowdh.bsky.social reposted
Dallas Card
@dallascard
Notes from the @FAccTConference (in person) town hall #facct22 #facct2022
-
ruchowdh.bsky.social reposted
The Tor Project
@torproject
So please, use Tor for anonymity when doing research about abortion access—we don't know how data collection will be used to harm people who even search for this information.
Download Tor Browser here: torproject.org/download -
ruchowdh.bsky.social reposted
Signal
@signalapp
Here is your friendly reminder that we built Signal for private, secure communication. It’s built so you can communicate individually and in groups, through text and calls, without fear of interference or data collection. Free to use and not for profit.
-
ruchowdh.bsky.social reposted
Erica Robles Anderson
@fstflofscholars
Strong media effects intuitions aren’t correct. Music doesn’t cause x, movies don’t cause y, video games don’t cause z, just as reading books won’t guarantee a, screen-free life won’t insure b, and computing won’t determine c.
-
ruchowdh.bsky.social reposted
Lea Kissner
@LeaKissner
PEPR! I'm not livetweeting today because my hands aren't up to it, but @tariqajyusuf is. We've kicked off with a whole session on differential privacy and I'm so happy to see more usage that solves real problems in the real world
It's not a magic wand, but it's useful math. -
ruchowdh.bsky.social reposted
wells (wants a better world)
@WellsLucasSanto
Liberty is not just positive liberty, or freedom *to* do whatever you want (in this case, an unlimited right to secrecy; in other cases, an unlimited right to speech). Liberty is also negative liberties, such as freedom *from* certain things (like surveillance, harassment, etc.).
-
ruchowdh.bsky.social reposted
wells (wants a better world)
@WellsLucasSanto
I still maintain that *so many* of our conversations today could be advanced by an understanding of "a right to" (positive liberty) and "a right from" (negative liberty). This isn't at all a new concept, and I'm glad that it's finally working its way back into the discourse.
The New Yorker
@NewYorker
Privacy actually encompasses two different concepts: the right to secrecy, and the right to be free of government interference. This has led to confusion over whether it really is a fundamental right. nyer.cm/gUkCwAl
-
ruchowdh.bsky.social
@ruchowdh
Having an idea and building something from an idea are two totally different things.
Thinking of something in the abstract is only a fraction of the work to be done in implementing it. -
ruchowdh.bsky.social reposted
Julia Angwin
@JuliaAngwin
In light of the recent US settlement with Facebook, I want to tell y’all a story about how hard it is to make change in our algorithmic world, why you need a village of researchers, and why law enforcement agencies need to get better at tech. /1
www.nytimes.com/2022/06/21/technology/meta-ad-targeting-s… -
ruchowdh.bsky.social
@ruchowdh
I'm a fan of @laughsovich 's blog in the current issue of @mtlaiethics newsletter - the team has done phenomenal work in applying inequality metrics to our assessments - here's a summary:
montrealethics.ai/measuring-disparate-outcomes-of-content… -
ruchowdh.bsky.social
@ruchowdh
Like defcon but not cool
-
ruchowdh.bsky.social reposted
dj patil
@dpatil
This is a great job/role in civic tech www.usajobs.gov/job/659818300
It's not for the faint of heart. It's a serious mission. Consular Affairs is responsible for citizen services (think: travel, passports, evacuations) and the issuance of nonimmigrant and immigrant visas worldwide. -
ruchowdh.bsky.social reposted
Mozilla Festival
@mozillafestival
Our last Trustworthy #ai Working Group was a success, thanks to Project Leads like @laughsovich, @KLdivergence and @ruchowdh who worked on visualizing internet subcultures on social media.
Bring your project to #mozfest for collaboration & feedback ▶️ mzl.la/taiwg -
ruchowdh.bsky.social
@ruchowdh
I just lived my life in these comments - from clearly articulated email strategy to someone earnestly explaining what a “carbon copy” is and hence a “blind carbon copy”
abby govindan
@abbygov
the day I found out that CCing someone has the exact same impact as just adding their email in the “send to” field is the day I realized all of adulthood and professionalism is just a facade
-
ruchowdh.bsky.social reposted
abby govindan
@abbygov
the day I found out that CCing someone has the exact same impact as just adding their email in the “send to” field is the day I realized all of adulthood and professionalism is just a facade
-
ruchowdh.bsky.social reposted
aaron is vibe shifting to mastodon
@binary_aaron
@KLdivergence presenting work from her, @zywind, and @amandarg__ about extending bias correction to large numbers of subgroups and scale well to our team’s problems #facct22 #facct2022
-
ruchowdh.bsky.social
@ruchowdh
Idk what I’m more proud of - twitter meta’s shitposting game, our meme game, or our headline grabbing tweets game
Luca (also on the other platforms)
@__lucab
i am insanely proud of this slide that describes what we do
Luca (also on the other platforms)
@__lucab
come to listen to @binary_aaron and I talk about how research and eng work together in our team
#facct #facct2022 #facct22
ruchowdh.bsky.social
@ruchowdh
I mean, how good is this thread tho -
aaron is vibe shifting to mastodon
@binary_aaron
in the beginning there was Research and there was Engineering.
They have always been at war. They will always be at war.
The war has been waged so long they no longer know how the war started.
Luca (also on the other platforms)
@__lucab
come to listen to @binary_aaron and I talk about how research and eng work together in our team
#facct #facct2022 #facct22 -
ruchowdh.bsky.social reposted
Luca (also on the other platforms)
@__lucab
i am insanely proud of this slide that describes what we do
Luca (also on the other platforms)
@__lucab
come to listen to @binary_aaron and I talk about how research and eng work together in our team
#facct #facct2022 #facct22 -
ruchowdh.bsky.social reposted
Anaelia (Elia) Ovalle
@ovalle_elia
CRAFT workshop to assess queer harms in AI was PACKED with insights driven by interdisciplinary teams. Thank you to all attendees. Tbh feel so grateful and honored to be here. Online session to follow. Let’s continue the dialogue. Fuerza~ #facct2022 @FAccTConference @QueerinAI
-
ruchowdh.bsky.social reposted
Mark Riedl
@mark_riedl
Erm no. I read the article. The graphic designer did a lot of iterations on the prompt, cherry picked an image, and then used a feature allowing portions of the image to be re-generated.
Cosmopolitan
@Cosmopolitan
PSA: The world’s smartest artificial intelligence (aka @OpenAI) made this magazine cover (yes, really!)—its first EVER.
-
ruchowdh.bsky.social reposted
aaron is vibe shifting to mastodon
@binary_aaron
in the beginning there was Research and there was Engineering.
They have always been at war. They will always be at war.
The war has been waged so long they no longer know how the war started.
Luca (also on the other platforms)
@__lucab
come to listen to @binary_aaron and I talk about how research and eng work together in our team
#facct #facct2022 #facct22 -
ruchowdh.bsky.social reposted
Luca (also on the other platforms)
@__lucab
come to listen to @binary_aaron and I talk about how research and eng work together in our team
#facct #facct2022 #facct22 -
ruchowdh.bsky.social reposted
Sarah Porter 🦋
@SColesPorter
WSAI agenda is looking hot 🔥
Update with content team - there were 18.6 “wow”s
If you’re not in Amsterdam this Oct at @WorldSummitAI & you work in #artificialintelligence then you need to get your skates on 🦦
Maybe just the year we tip the scales as the world best tech event -
ruchowdh.bsky.social reposted
Liz Bright Marquis
@lizbmarquis
Happening now: CRAFT session on imagining meaningful collaboration between companies and communities to redress algorithmic harms. Super excited to learn from @Combsthepoet, @hypervisible, @data_beth, and Fabian Rogers 🔥🔥
Liz Bright Marquis
@lizbmarquis
Attending @FAccTConference virtually this week! While I'm hoping to take in as much of the conference as possible, I'm very excited to be involved in 2 sessions on algorithmic harms and accountability.
-
ruchowdh.bsky.social reposted
Amber Sinha
@ambersinha07
I recently began a fellowship with @mozilla to work on #aitransparency and #airegulation. It is an honour to be part of such an illustrious cohort and I look forward to working with them @AprylW @NeemaIyer @bobirakova @followlori @bgeurkink @Abebab foundation.mozilla.org/en/blog/new-fellows-start-work-to-…
-
ruchowdh.bsky.social reposted
wells (wants a better world)
@WellsLucasSanto
This is tomorrow!!! Everyone (and I mean *everyone*) involved as panelists, moderators, and organizers are incredible. I'm sad that only half of it is online, but I look forward to (virtually) seeing y'all there!
Be sure to apply for the session via the link in the thread!
Irene Font Peradejordi 🎈
@if_peradejordi
Join @Kyra_Yee, @laughsovich, @lizbmarquis, and I at @FAccTConference to discuss collaboration opportunities between companies & communities to remediate algorithmic harm. Panels will be moderated by @ruchowdh and @KLdivergence. Link to apply & panelists in the 🧵
-
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Food in Seoul 🧵
1. 가로수길칼국수, Kalguksu restaurant in Sinsa-dong. Korean auntie kimchi is unbeatable -
ruchowdh.bsky.social reposted
Meg Young
@megyoung0
hi #facct22 here is our paper facctconference.org/static/pdfs_2022/facct22-110.pdf the tldr is that medicine and public health have lessons for us on COI and we need to align with advocates and activists
-
ruchowdh.bsky.social reposted
Luca (also on the other platforms)
@__lucab
heading to #facct2022, my first in person conference since the one in 2020
who’s there and whom should i talk to? -
ruchowdh.bsky.social reposted
Jiahao Chen @NeurIPS
@acidflask
Travel tip for @FAccTConference : the Intercontinental Seoul COEX hotel is walkable from Bongeumsa 봉은사 奉恩寺 station on line 9 (local or express). Take exit 7, walk past the stairs on the right and take the mall entrance to the hotel lobby past the aquarium.
-
ruchowdh.bsky.social
@ruchowdh
TIL that “good luck to you” is Aussie for “bless your heart” and I am here for it. What others have I missed?
-
ruchowdh.bsky.social reposted
VCs Congratulating Themselves 👏👏👏
@VCBrags
There are too many angel investors
Where are all the demon investors? -
ruchowdh.bsky.social
@ruchowdh
The entire racket built around filing taxes in the US.
Mima
@LizzieCosmos
What’s a scam that’s become so normalized that we don’t realize it’s a scam anymore?
-
ruchowdh.bsky.social reposted
Tristan Snell
@TristanSnell
So the guy who flirted with buying this site to 'defend free speech' just fired a bunch of his spaceship employees for speaking out to criticize his behavior and failing leadership? Got it.
-
ruchowdh.bsky.social reposted
Bobi Rakova
@bobirakova
Fascinating to engage with the @OxConnectedLife @oiioxford community in presenting our work on exploring a new feminist-inspired social contract for how we come to terms with algorithmic systems with regards to bringing improved transpacrency to AI 📢 termsweservewith.org 🧵
-
ruchowdh.bsky.social reposted
Coty Rosenblath
@Coty
We should all follow @dionnewarwick's lead:
-
ruchowdh.bsky.social reposted
Rua M. Williams
@FractalEcho
I am so excited to be a #justtech fellow (@ssrc_just_tech @ssrc_org) with @BlackDigitalHum, @hypervisible, Christine Miranda, @ClarenceOkoh, & @wemeasureus. I'm excited to talk about my project #cyborgcoalitions. But first, I want to celebrate my new colleagues. 🧵
-
ruchowdh.bsky.social
@ruchowdh
Whoever did this? I’m disgusted to be at the same company as you and hope we’ve never met nor will ever meet.
rat king 🐀
@MikeIsaac
the twitter all-hands today was certainly leaky, but the twitter employee(s) leaking to Johnson know exactly what they’re doing:
Johnson now blasting out video/tweets with all of the names of employees pushing back against Elon -
ruchowdh.bsky.social reposted
@timnitGebru (@dair-community.social/bsky.social)
@timnitGebru
Thinking about @ruchowdh @KLdivergence @ubiquity75 and the META team that Rumann built at Twitter. People do not understand how much fighting it takes to create remotely helpful in these companies to move the needle and one man is about to come in and destroy everything?
-
ruchowdh.bsky.social
@ruchowdh
Twitter twitter never misses
-
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
This is going to sound funny but Musks immediate chilling effect was something that bothered me significantly. Twitter has a beautiful culture of hilarious constructive criticism, and I saw that go silent because of his minions attacking employees.
-
ruchowdh.bsky.social
@ruchowdh
-
ruchowdh.bsky.social
@ruchowdh
The latest from META — an analysis of algorithmic amplification of tweets from elected officials in Canada, France, Germany, Japan, Spain, UK, & US.
The team: @fhuszar, @s0f1ra, Conor O’Brien, @__lucab, @sagemintblue @mrtz. 1/n
blog.twitter.com/en_us/topics/company/2021/rml-politicalc… -
ruchowdh.bsky.social
@ruchowdh
I had two seconds between meetings to post my last post so I’m going to take some time here in appreciation and awe of @safiyanoble - she has repeatedly set the bar, as an academic, as a friend, and as a generally real and good human. There is no one who deserves this more 1/
-
ruchowdh.bsky.social
@ruchowdh
Doing the happiest of happy dances!!!
-
ruchowdh.bsky.social
@ruchowdh
I'm about to be live with @kaifulee to discuss his latest book, AI 2041, hosted by @cwclub
youtu.be/GrHKHvikRZo -
ruchowdh.bsky.social
@ruchowdh
It worked so well on children, let’s enforce arbitrary and punitive invasive and incorrect technology in the workplace too!!
This is NOT the takeaway to have.
www.wired.com/story/ai-can-run-work-meetings-now-headroom… -
ruchowdh.bsky.social
@ruchowdh
I have to admit, I got tired of trying to fight with Cade and the NYT. It’s hard to explain how draining it is and how stupid you feel having big institutions gaslight you.
Thank you @histoftech for writing this and officially correcting the record.
www.technologyreview.com/2021/08/03/1030460/women-erased-… -
ruchowdh.bsky.social
@ruchowdh
It's up! Would love thoughts from the community on our attempt, and encourage people to sign up and participate!
Engineering
@Engineering
Calling all bounty hunters - it’s officially go time! We’ve just released the full details of our algorithmic bias bounty challenge which is open through August 6. For more details on the challenge, head over to our blog 👇 blog.twitter.com/engineering/en_us/topics/insights/2021/a…
-
ruchowdh.bsky.social
@ruchowdh
We are launching the first ever algorithmic bias bounty tomorrow at 9am PT! Learn more about it and hear from our judges at our spaces event tomorrow afternoon -
x.com/i/spaces/1OdJrVzeQykJX -
ruchowdh.bsky.social
@ruchowdh
It's been hard to keep quiet on this ... excited to announce Twitter META will be hosting the first *ever* algorithmic bias bounty competition @aivillage_dc @defcon. Details of the challenge coming later this week. In the meantime...
-
ruchowdh.bsky.social
@ruchowdh
Why I’m skeptical - a language model assessing media is a secondary source. It will not, and cannot, replace primary source work. It’s the algorithmic equivalent of getting your news from a talking head; you need to realize you are getting a version of reality. 1/
MIT Technology Review
@techreview
The idea is that instead of searching for information in a vast list of web pages, users would ask questions and have a language model trained on those pages answer them directly. www.technologyreview.com/2021/05/14/1024918/language-mode…
-
ruchowdh.bsky.social
@ruchowdh
Reminder that we are hiring for the ML Ethics team at Twitter! I’m looking for a Staff/Sr. Staff Engineer and an Engineering Manager. Job postings:
careers.twitter.com/en/work-for-twitter/202101/f59d2933-d… -
ruchowdh.bsky.social
@ruchowdh
Responsible ML isn't a team, it's a company-wide imperative. With META as the driver, we're excited to share our new RML initiative. 1/5
Engineering
@Engineering
Today we’re introducing the Responsible ML initiative, our effort to understand the use and impact of machine learning at Twitter, and take action when needed.
We’ve been working on this for a while. Here’s how we see the path forward:
blog.twitter.com/en_us/topics/company/2021/introducing-re… -
ruchowdh.bsky.social
@ruchowdh
.@williams_jutta is guiding force on META, looking forward to this panel!
-
ruchowdh.bsky.social
@ruchowdh
Big news! - I’m thrilled to be joining the @TwitterEng team today as Director of ML Ethics, Transparency & Accountability. With the META team, we’ll work to improve ML transparency, inclusivity and accountability. 1/
-
Post not available here, view on X
ruchowdh.bsky.social
@ruchowdh
Absolutely agree w your take - we are the optimists, because we know we can do better with technology. We spend a lot of time and energy working WITH tech leaders to advise, build and improve. We want everyone to have the advantages and we want to face the harms head on.
-
ruchowdh.bsky.social
@ruchowdh
“80 to 90 per cent of the predictive assessment was based on the algorithms’ analysis of candidates’ use of language and verbal skills.”
By design, excluding non-native language speakers. Disturbing given the current nationalistic wave in the UK.
www.telegraph.co.uk/news/2019/09/27/ai-facial-recognition… -
ruchowdh.bsky.social
@ruchowdh
By me and @gleemie on tech’s quiet colonization of the social sciences:
“The beautiful messiness of humanity cannot be quantified into data columns, programmed, scaled, and engineered.”
www.wired.com/story/tech-needs-to-listen-to-actual-resear… -
ruchowdh.bsky.social
@ruchowdh
Wait until someone tells them how averaging works. www.theverge.com/2019/5/29/18644143/uber-deactivate-rider…
-
ruchowdh.bsky.social
@ruchowdh
I encourage any data scientist from an underrepresented group to apply. If you’d like, I’m willing to help you create a strong application.
Matthew Salganik | @msalganik.bsky.social
@msalganik
The @nytimes data science group has 3 open positions. This is a chance to be part of a great team doing interesting and important work. See below for more about each position. Application review is happening now.
-
ruchowdh.bsky.social
@ruchowdh
We cannot divorce 'making things work' and 'impact on society' when it comes to applied artificial intelligence. Frankly, your AI does not "work" if it is biased, perpetuates social inequality and discrimination, or reinforces unequal power structures.
www.forbes.com/sites/cognitiveworld/2018/12/20/geoff-hint… -
ruchowdh.bsky.social
@ruchowdh
Ethics and AI: A Beginner's Guide is updated! Added a doc of academic syllabi (thanks @InternetEthics) docs.google.com/document/d/1xEhKmlcrvdssA-yoCDK035GvwYmOO…
-
ruchowdh.bsky.social
@ruchowdh
International #datascientists - @thisismetis now provides assistance with M-1 visas for our bootcamp in New York!
-
ruchowdh.bsky.social
@ruchowdh
Trying to get hired in data science? What are the skills you need beyond programming and stats?
www.rummanchowdhury.com/readme/2016/7/11/how-i-got-into-d…















































































































































































































