Penn State's Philosophy Department is looking for an AI ethicist! Please share this ad with everyone who might be interested. philjobs.org/job/show/320...

Brilliant

Samuli Reijula@samulireijula.net · 3d ago

"i am, myself, a lossy compression" by @henryconklin.bsky.social , very cool! henryconkl.in/posts/i-am-a... #cogsci

I wrote an e-mail to James Hollis recently (few months ago). Seems frivolous...e-mail. I should have sent a handwritten note. I was longing for some way to connect with this absolute giant in my life. (remarkably) I received a response the next day. With a pearl of wisdom I'll always cherish.

Béa Gonzalez@sophiacycles.bsky.social · 6mo ago

Whenever one goes through the deconstruction of the false self-one normally suffers a considerable period of wandering in the wasteland.~James Hollis

Just got done listening to the @404media.co podcast interview with Sherry Turkle. Lots of wise stuff in there about empathy and how artificial friends aren't just a poor substitute for actual human connection but really harmful. >>

I *hate* this "digital divide" framing for AI; even libraries marshal it to justify their "AI literacy classes," ensuring all have "access" for self-actualization and economic opportunity. Maybe instead some folks are sufficiently literate, and they're instead choosing *principled refusal*.

The New York Times@nytimes.com · last wk.

The Gates Foundation has pledged to spend at least $1 billion over the next two years to expand global access to A.I. and help “narrow gaps between the richest and poorest rather than widening them.”

Big (huge?) news on the computational complexity preprint server, with a paper by Yumou Fei, Dor Minzer, and Shuo Wang I am not qualified to read beyond the intro: the perfect completeness version of the (not) Unique Games Conjecture, the 4-to-1 Games conjecture! eccc.weizmann.ac.il/report/2026/...

We could waste a lot of time speculating about the apocalyptic worldviews of tech billionaires. Whatever they believe, talk of existential risk is a convenient way to not address known harms produced by AI systems overseen by fascist billionaires. @theintercept.com theintercept.com/2026/09/14/a...

Tech CEOs’ Doomsaying Is a Distraction from Real, Existing Harms of AI

The binary between AI apocalypse and AI redemption asks us to ignore data centers and AI war machines — all as tech executives get incredibly rich.

theintercept.com

cloak and dagger navier stokes cims.nyu.edu/~tristanb/st...

Co-evolution, ecology, and the absolutely brilliant Michael Levin 😍 arxiv.org/pdf/2609.03344

OpenAI is now facing more than 50 consumer harm and wrongful death lawsuits alleging that extensive use of ChatGPT resulted in psychological harm, physical injury, and even the deaths of users and/or those around them. futurism.com/artificial-i...

OpenAI Is Now Facing Over 50 Consumer Harm and Wrongful Death Lawsuits

OpenAI was hit with 30 new lawsuits overs its alleged role in the Tumbler Ridge school massacre. It now faces over 50 consumer harm lawsuits.

futurism.com

👀 and 🤍 for Michael Zhang tomorrow.. he has our superMario in his corner www.atptour.com/en/news/anci...

"Trying to sell us a solution to a problem they created" www.youtube.com/watch?v=Rwv8...

What Really Happened When OpenAI Bots Escaped a Cybersecurity Test? | Amanpour and Company

YouTube video by Amanpour and Company

youtube.com

You're not the only one finding it hard to cancel your Amazon Prime subscription. Apparently the 4-page, 6-click, 15-option cancellation workflow is nicknamed the "Iliad" internally. (16'30'') www.youtube.com/watch?v=Wko-...

I love the dedication in My Life on the Road by Gloria Steinhem

The most thoughtful position I've seen on AI from any University thus far. @mit.edu aiandeducation.mit.edu/report/#23-p...

🚨 JCDL 🚨 To effectively aggregate, reason over scientific literature, we need ways to extract key info (e.g., hypotheses, statistics) from published papers (PDFs... not ideal...). We tackle ways to do that (better) using LLMs. We conclude this is still a hard problem. arxiv.org/pdf/2603.21193

For all the things I get wrong, this is one I think I get right. When I think someone is extraordinary, I tell them. About the little moments, their words that stayed with me, their beautiful expressions. (we need) More kindness, more connection.

Will Oremus@willoremus.com · 4w ago

randomly giving out genuine compliments is such an easy way to make the world a little nicer, i'm always surprised more people haven't figured that out and have a ton of respect for those who have

🚨 ACM GROUP 2027 🚨 Another labor of love story... years in the making and revising. We interviewed 25 individuals from marginalized and vulnerable groups who turned to online community for support when offline resources were scarce. preprint (although, older version): arxiv.org/abs/2311.15427

Lived experiences of online harm amongst marginalized and vulnerable individuals in support-seeking communities on Reddit

Online communities can serve as meaningful sources of social support, particularly for marginalized and vulnerable groups. Disclosure of personal information facilitates integration into online commun...

arxiv.org

Afghan women leave their house 1-2x per month. They are forbidden to go to school after 6th grade. Profs: 200 volunteer professors can help you repurpose your existing class, in English, for these deserving girls. AFSO students are being admitted to universities and some of them are getting out. 🤍

Afghan Female Student Outreach | University education for Afghan women

Afghan Female Student Outreach (AFSO) builds pathways for Afghan university women to complete their education

afsousa.org

🚨 EMNLP x2 🚨 2 papers accepted to EMNLP 2026 1. **Affective Context Amplifies Sycophancy in LLM Responses** - title says it all.. 2. **Aggregation-Aware Synthetic Text Generation Against Authorship Re-Identification** - privacy-preserving synthetic data generation outperforms SoA

Friction

Sung Kim@sungkim.bsky.social · last mo.

Mathematics in the age of AI by Terence Tao An essay on how the mathematical community might respond to the arrival of AI tools that are capable of performing research-level mathematical tasks. arxiv.org/abs/2608.16753

Acceptable plan for sabbatical when you’re a full Professor (apparently) #elcap

"The Analytical Engine weaves algebraical patterns just as the Jacquard-loom weaves flowers and leaves." -Ada Lovelace, 1843 (computing was always women's work)

💔

Croatia Week @croatiaweek.bsky.social · last mo.

As flames closed in, people took to the sea. 194 rescued. More about the rescue operation on the Dalmatian coast here: www.croatiaweek.com/croatia-wild...

A new low. www.wired.com/story/meta-r...

Meta Ran Ads That Contained AI-Generated Child Sexual Abuse Imagery

More than 50 offending image and video ads were published across Facebook, Instagram, Messenger, or Threads, according to Meta’s ad library data. Some ran as recently as this week.

wired.com

Yes. Have been thinking this way too re:AI.. Digging into hydrothermal vents and similar.. But is it just an apt metaphor or can we map it more closely?

Louis Redux@louisredux.bsky.social · 3mo ago

An environment where, largely removed from the necessity of the right molecule in the right place, information itself is the wave of "experiments" being crashed against the deep sea thermal vents/tide pools etc of in silico evolution. Life or not, self-replicators will evolve. 7/n

🚨New paper - out today in Vaccine 💉 papers.ssrn.com/sol3/papers.... Our findings illustrate how the engagement architectures of social media platforms shape elite political communication and the informational politics of public-health governance.

“I am no bird; and no net ensnares me; I am a free human being with an independent will.” 💌

“It took the world of sampling to the next level. … It wasn’t so much the actual samples they used as the way they aligned them together into a seamless flow. It was the epitome of a masterpiece.” Inside the Making of Beastie Boys’ ‘Paul’s Boutique’:

“It was the epitome of a masterpiece…”

The Beastie Boys, famous fans and collaborators recall the creation of the rap trio's 1989 sampledelic game-changer, Paul’s Boutique.

mojo4music.com

Can we please stop pretending AI can replace peer review? If I wanted my paper reviewed by a model, I could have done it already. I want my community to read it and provide feedback. That's the whole point of peer review. Science depends on expert scrutiny and accountability—not automated judgment.

Emily M. Bender@emilymbender.bsky.social · 2mo ago

I did not submit anything to #EMNLP2026, which is probably part of why I missed this when it was first posted in May, but I want to say now that I am utterly appalled that @aclmeeting.bsky.social & @emnlpmeeting.bsky.social are engaging in this nonsense. 2026.emnlp.org/ai-reviewing... A thread 🧵>>