Got a pretty hilarious warning from Google T&S today @defcon.bsky.social @aivillage.bsky.social There’s actually nothing that should flag this, it’s just screenshots, images, and text and none of it is about phishing. My talk was about model stealing though APIs.

On Wednesday, the Senate Commerce Committee will vote on four bills that would expand age verification, increase online surveillance, and make it harder to access lawful speech online. Take one minute to tell your senators to vote NO. www.acteff.org/campaign/16...

Tell the Senate: Don't Turn the Internet Into an ID Checkpoint

Tell the Senate to Reject Bills That Expand Age Verification

acteff.org

These concerns are very much not hypothetical. In 2024 a private benchmark called FrontierMath developed by a third party called Epoch came out. We later learned that OpenAI funded the creation of the benchmark and while nobody else had the test Qs, OAI did. ai-frontiers.org/articles/don...

Don’t Let AI Developers Hire Their Own Referees | AI Frontiers

Gabriel Weil, Jul 29, 2026 — Letting AI developers pick their own safety auditors creates a conflict of interest. Requiring liability insurance instead would put insurers’ own capital behind risk asse...

ai-frontiers.org

Remember, only Chinese models can protect you against a cyberattack from OpenAI

cbass@stochasticbrainrot.com · 2w ago

>be me, sam altman >announce evil version of AI for cyberwar >inadvertently attack open model site >guy I hired yells in public abt dangerous chinese AI is >hacked org can't secure themselves with US AI (isn't a member of the elect), has to turn to chinese AI >have to admit role in attack >mfw

If you believe their story, OpenAI accidentally committed cyber warfare against Hugging Face while doing what they thought was an internal test of a model without internet access. In six months time will be obvious they’re faced effectively no sanction for this.

This is a illegal. The admin cannot do this. This is also deeply immoral: Trump is killing people because he doesn’t like their governors.

Catherine Rampell@crampell.bsky.social · 2w ago

Trump administration pauses $1 bln in Medicaid payments to California, Minnesota www.reuters.com/world/us-hea...

Whenever I interact with an academic discipline outside of AI and Theoretical Computer Science, I am rapidly reminded how few problems I face for being trans in AI. Thank y’all <3

When I was in middle school (late 2000s), I was very interested in mathematics. I thought that the validity of the proof of the 4-color theorem was disputed in mathematics and was shocked when I learned about FLT, but I knew that humans would never beat an AI at chess ever again

Today I met someone at arXiv who said they were a huge fan of the Pile. I assume at some point one gets used to your idols being fans of your work, but six years in I still haven't.

An updated guide to where you can find @eleutherai.bsky.social at #ICML2025! We have a lot going on, including an oral presentation tomorrow (Tuesday) where I’m going to be talking about my scientific research agenda. Come say hi to us!

Claude code is stenographically marking location information in its prompt to help Anthropic profile users by where they (appear to) be located thereallo.dev/blog/claude-...

Claude Code Is Steganographically Marking Requests

I inspected Claude Code for privacy reasons and found hidden system prompt markers based on API base URL and timezone.

thereallo.dev

I’m walking back from an AI event to my hotel and I look up and see a rock climbing gym called “Benchmark.” Is this what living in SF is like?

This is a crazy good article about AI and environmental impacts: blog.andymasley.com/p/a-cheat-sh... By @andymasley.bsky.social

Using ChatGPT is not bad for the environment - a cheat sheet

The numbers clearly show this is a pointless distraction for the climate movement

blog.andymasley.com

Infrasound harm from data centers make it to the NYT. Unless the data center's sound is so loud that you can feel your body physically shaking like you're at a concert, I don't think this has any impact on your health at all. Been crazy to see this idea take off. www.nytimes.com/2026/06/17/u...

I don't think "urgent situation" includes an event you literally had 250 years of advance notice to plan ahead for.

Philip Bump@pbump.com · 2mo ago

I’m sorry, no. This is not the guy who did the reflecting pool. No. Come on. We’re just putting a hat on a hat here. www.nytimes.com/2026/06/18/u...

The ability of tech co. to produce "intelligent" systems that are incompetent at doing any of the tasks I actually want them to do is mind-boggling. TIL that ChatGPT and Claude generally don't agree when you give them two papers and ask how many citations they have in common.

In film, "we'll fix it in post" is what you say when something went wrong on set and you don't want to redo it. AI research has made it our entire methodology: train the model, then patch whatever comes out. Our new ICML oral argues this can't be the basis of a science of AI. 🧵

I had given Anthropic a lot of credit for turning down the DoD and its trillions of dollars, especially as it seemed to be the only example of any AI company making any financial sacrifices for moral principles. Of course, it turns out to be not really true. www.axios.com/2026/04/19/n...

Scoop: NSA using Anthropic's Mythos despite Defense Department blacklist

The government's cybersecurity needs are outweighing the Pentagon's feud with Anthropic.

axios.com

"[W]hen these goods remain concentrated in the hands of a few, without adequate forms of sharing and access, a new imbalance is created that contradicts the universal destination of goods" Very cool to see the Pope endorsing @eleutherai.bsky.social's mission

Congrats guys! Racism is over, so now it’s legal to draw congressional districts to systematically disenfranchise black people.

My hot take is that the median social system breaks under too much optimization pressure, and we should stop trying to optimize things

Ethan Mollick@emollick.bsky.social · 4mo ago

Every system that was regulated, either explicitly or implicitly, by the fact that they were effortful for humans (letters of recommendation, government filings, essays, or, as this paper finds, lawsuits) will break under a wave of AI.

Excited to be on my way to @iclr-conf.bsky.social! Come stop by our posters and hit me up. I'm especially excited to talk about - Open weight safety - Training dynamics and interpretability over time - Memorization and machine unlearning - Open data - Rigorous experimental design

@eleutherai.bsky.social · 4mo ago

Looking for EleutherAI @iclr-conf.bsky.social? Come by our posters! If you're in our discord, we have a thread #general > ICLR 2026 Meetup you can join to coordinate with @stellaathena.bsky.social, Goncalo Paulo, @norabelrose.bsky.social, and members of our community who will be there!

FISA 207 is blatantly illegal and immoral and has always been obviously so. Republicans are pretending to not know this, just like Democrats did during the Biden administration. This is bipartisan evil.

Feb 3, 2025 - We started fighting to save our data. July 3, 2025 - We launched #SaveOurSigns with Minn librarians. April 2026 - We are still talking about the importance of public data as a public good. ❤️🛟

National Security Archive@nsarchive.bsky.social · 4mo ago

Today, the Archive published a Disappearing Data Chronology--a timeline tracking changes in access to federal information under Trump, including major data losses and restorations, legal challenges to information takedowns, and threats to archival collections. nsarchive.gwu.edu/special-exhi...

Regretfully, the story about LLMs anti-polarizing people was not real.

Alexander Hoyle@alexanderhoyle.bsky.social · 4mo ago

Apparently it's based on LLM-simulated users (and LLM judges). I generally think highly of the FT's reporting but this is is nonsense

If this is real, it’s very plausibly the biggest win for alignment research.

Dare Obasanjo@carnage4life.bsky.social · 4mo ago

Testing 61 policy questions, John Burn-Murdoch's FT analysis found that major AI chatbots consistently pull users away from fringe views. Grok nudged responses center-right, while GPT, Gemini, and DeepSeek pulled them center-left. This is compared against social media where extreme views dominate.

You have a moral imperative to refuse to work with these people or develop models for these purposes.

How do you identify which problems are interesting and valuable? When people don’t work on problems that matter, why do you think that is?

If I was going to claim that a finetuning methodology for machine unlearning “really worked,” what evidence would you like to see?