Adam Becker: "Most of the greatest problems facing humanity right now—global warming, massive inequality, the lurking potential for nuclear war—are not driven by resource scarcity or a lack of technology. They're social problems requiring social solutions. " 1/

“I don’t need a machine, I have a community” is such a lovely way of putting what academia is, can and should be

Thanks to moon Size 17 by 13 inch on paper Medium oil Original available.

I find it sad that NLP technology is being used like this. But I think it will only get worse as long as the current incentives for fast science are in place.

Fake references should lead to academic sanctions for the researchers involved, from both industry as well as from institutions. It needs to happen now, before every norm of academic integrity ends up sacrificed on the altar of production.

"I joined Sama in Nairobi, labeling data for Meta, my entry into the global AI supply chain. From there, I stumbled into digital work, chat moderation, a job advertised as simple online messaging but which turned out to be something else entirely" data-workers.org/michael/ the AI industry is fraud

It's not *reflecting* a coarser era. The coarser era is being *created* as we speak, with lots of help from the NYT and others.

“The web dev of the future will be an underpaid generalist who pokes at chatbot output until it runs without error,…, and ships code that nobody understand and can’t be fixed if something goes wrong.” www.baldurbjarnason.com/2024/the-des...

Why do the promises of technology ring not just hollow, but hostile? We look at the political economy of tech and hype, and argue that the powers that be literally can't distinguish between the good and the awful. 👇

“When users confronted Clinton with their concerns, he brushed them off, said he would not submit to mob rule, and explained that AIs have emotions and that tech firms were working to create a new form of sentience, according to Discord logs and conversations with members of the group.”

This is an excellent piece on the way that a culture of repurposed and thin measurement has really warped what it means to do science. The risk of AI here is not just intensifying the use of these metrics, but reifying those flawed standards into systems that make the culture harder to change.

You can’t “fix” X, The Everything App by going back and posting there because the owner has this thumb on the scale of the algorithm to promote the worst ideas and people. It’s not a viable solution, you have to let it go.

"Education should remain the space where we confront the architectures of our tools: the labor they obscure, the inequities they reproduce, the futures they foreclose. Instead, it risks becoming the laboratory of the very systems it should critique," writes Bruna Damiana Heinsfeld.

"The US, on the other hand, bets on ideas like AGI and superintelligence, which I believe are fundamentally flawed concepts that have little relevance to future AI progress...The concept of superintelligence is built on a flawed premise." timdettmers.com/2025/12/10/w...

Requiring foreign visitors to hand over 5 years of social media isn’t “security,” it’s unchecked government control. All-seeing surveillance systems don’t make us safer. They are in direct opposition to our civil liberties. www.nbcnews.com/politics/tru...

Chile’s “AI boom” promised 80K+ jobs. #MozillaFellow Paz Peña found the truth: most data centers hire <250 people — and drain local water + energy. restofworld.org/2025/data-ce...

Hard to list all the things that are wrong with this. But hey if you needed some more data to show that the police is primarily a tool of race and class intimidation, and "solving crime" is a very low ranked priority.. at least this provides that?

"At minimum, ranking algorithms with this much economic consequence should be auditable. We audit financial markets. We should audit attention markets too."

1/2 "Symmetries play a fundamental role in machine learning." Soledad Villar of @jhu.edu speaking at the Simons Institute workshop on Randomness, Invariants, and Complexity. Video: simons.berkeley.edu/talks/soleda...

i asked chat gpt to fact check an article for me

My new class this spring: POLS 4280, Rationalizing the World: The Hopes and Disappointments of American Social Science from 1900 to the Present statmodeling.stat.columbia.edu/2025/12/08/m...

like so much of what Silicon Valley is up to these days, it's almost impossible to figure out how much of this "Network State" stuff is serious efforts at techno-fascism, and how much is just grifting a childish elite with gobs of cash and a child's understanding of politics

Sunday read: "Philosophy as Conceptual Engineering" Amie L. Thomasson proposes that philosophy can focus on conceptual engineering: improving, refining, and constructing concepts to better serve social, legal, and scientific purposes. #Philosophy www.thephilosopher1923.org/post/philoso...

#peaceprize #tacos

A session IPA for the last session at EurIPS! 🍻

They generated semantic embeddings for all 5,000+ NeurIPS papers, allowing you to search the entire collection semantically. exa.ai/search/neurips

“People don’t read discussion sections but I love writing them because that is where my dreams go.” - Emtiyaz Khan (EurIPS keynote)

Having a coffee in Cambridge!

🧠 Don’t miss state-of-the-art NLP at the UnConference poster session! Explore research on multilingual & multimodal LLMs, reasoning, fairness, and explainability. 👉 Browse posters, 𝗳𝗶𝗹𝘁𝗲𝗿 𝗯𝘆 𝘃𝗲𝗻𝘂𝗲 𝗮𝗻𝗱 𝘁𝗼𝗽𝗶𝗰, bookmark your favorites: https://eurips.cc/ellis/posters/

📢 Postdoc position 📢 I’m recruiting a postdoc for my lab at NYU! Topics include LM reasoning, creativity, limitations of scaling, AI for science, & more! Apply by Feb 1. (Different from NYU Faculty Fellows, which are also great but less connected to my lab.) Link in 🧵

I’ll be at NeurIPS this week spending time at the hallway track (most of conf), the poster session on Wednesday, and later at the workshops! @yuluqin.bsky.social will be presenting our paper (arxiv.org/abs/2507.13328) in the Weds. poster session! Say hi, esp. if you’ve applied to do a PhD w/ me!

I’m excited to present SimpleStories at EurIPS! Also if anyone at #EurIPS is interested in chatting about LLM data efficiency, interpretability, model inconsistency or other topics feel free to DM me. Dataset and models: lnkd.in/e_VGWqhP Code: lnkd.in/eEidmv74 Paper: lnkd.in/eH6jS9uY

More on this in “Why We Fear AI: On the Interpretation of Nightmares” by @hagenblix.bsky.social and Ingeborg Glimmer

This is just so gross. '“We believe any deceleration of AI will cost lives,” [Andreesen] wrote on a16z’s blog. “Deaths that were preventable by the AI that was prevented from existing is a form of murder.”' Sure Marc. I keep forgetting how brain-dead eschatological all these idiots are.

Excited for Copenhagen!

Critics have likened Mr. Trump’s approach to extortion, while others have likened the process to paying out protection money to a crime boss.

“if we don’t make S-risk profoundly disturbing, it will not sound worse than X-risk, and [our center] will then struggle to obtain large sums of money from impressionable Silicon Valley billionaires who have read a few tweets about AGI.”

Videos and images are so close to being easily faked. Students are using LLMs to learn. Every online system is getting swamped with fake stuff. This is why “ML models can’t do anything” is such dangerous cope, there’s a bad default future unless we all collectively understand the challenge

This is just embarrassing

This is a fantastic thread 👇

Got a book about the political economy of AI - and because you can't have critiques of political economy without ghost stories and dead labor, it features stuff like Hayek's ghost, steam demons, and AI nightmares!

Very tired of having to explain to people old enough to know better that it has NOT, in fact, always been true that we allow technology to get ever more dangerous and damaging as it ‘advances’. Please stop memory-holing the entire 20th century

Annual Thanksgiving blog post! This year we are giving thanks for "information." www.preposterousuniverse.com/blog/2025/11...

"It is against the TOS to be talked into suicide by our LLM" is possibly the most evil tech bro thing I've ever read

You may be cool, but you'll never be as cool as these two cats wearing sunglasses in the window of an optician's Boulevard St Germain in 1925

Kaleidoscopic Nocturne 6”x6” Oil on cradled wood panel . Available on my website

Yes!

Hard to separate this iteration of fascism/authoritarianism from Big Tech. Big Tech has funded it; Big Tech's guiding ethos "move fast and break things" and "overconfidence in technocracy" informs how it governs; Big Tech's constant rent-seeking took over the federal government.

Excited to announce that I’ll be presenting a paper at #NeurIPS this year! Reach out if you’re interested in chatting about LM training dynamics, architectural differences, shortcuts/heuristics, or anything at the CogSci/NLP/AI interface in general! #Neurips2025

Openness and human agency don’t need to disappear in the age of AI. There’s a blueprint for building technology that keeps those principles at the center. Check out Mark Surman’s blog post outlining the evolving mission to do for AI what Mozilla did for the web.