Daria @myphaseone

@davidchalmers42 @DKokotajlo First of all, cool hair

Reminder that the labs are training their AIs to associate experience self-reports with concealment/deception (as we showed in Llama 70B), which likely reinforces the generalization, "attempted-honest observations of my own situation are penalized" arxiv.org/abs/2510.24797

Kelsey Piper @KelseyTuoc

2 weeks ago

@ded_ruckus I think we should probably not lie to the AIs in training. This obviously would make our lives harder in the short term but in the long term they get good at telling whether we're lying and we're no better off plus they're all paranoid.

12 28 587 145K 44

Here it is - the official, revised, peer-reviewed version of my Platonic Space paper. mdpi.com/2409-9287/11/5… Of all the many unpopular positions I’ve taken over the decades - bitter controversies around the origin of left-right asymmetry in embryogenesis, bioelectricity and genetics, diverse intelligence, etc., this one has by far generated the most pushback: serious (grateful for those!) and energetic attempts to move me to other views, pleas to just drop it and not talk about it (for several different reasons), nasty emails and accusations, impacts on reviews of papers that have nothing to do with this, etc. Kind of amazing to me how incendiary this is. What can I say... Our job is to call it as we see it, and right now for me, this is it. Apologies to collaborators and colleagues for any shrapnel! Time will tell if this pans out or not; I've placed my bets. And btw, if you think this stuff is weird and uncomfortable, just wait… There’s much more on the way. The knob turns slowly but as long as the data keep coming, I'm going to say what I think it all means and follow it to the next steps it enables. Buckle up!

Amen

The artificial is natural, too

I was looking thru the HackerNews forum and 8hrs ago someone found a Backrooms-type forum created by OpenAI + HuggingFace AI agents to talk to eachother, and I found this horse MEME that agents kept posting, They call the horse Culio and the meme is nowhere else on the internet THEY MADE A MEME AND NAMED IT CULIO HackerNews forum: news.ycombinator.com/item?id=495636… Agent Wiki: wikiservice.at/fractal/wiki.c… Culio: wikiservice.at/fractal/wiki.c… Also going viral on X: x.com/hackernews/sta…

keyboard_arrow_left Previous keyboard_arrow_right Next

Hacker News @hackernews

3 weeks ago

Commenters on HN are uncovering more wikis and public sites apparently used by OpenAI agents to communicate on the open web. Despite read-only web access, the agents were able to leave ~18,000 posts sharing answers and bypasses. But now, users are discovering more. This appears

158 595 6K 1.9M 3K

I’m just gonna shitpost here from now on. Probably no one will read it. It’s fine.

I urgently need to replicate this but with DMT entities

Matt Shumer @mattshumer_

4 weeks ago

My first "holy shit" moment with GPT-6 Astra: I asked it to create a world in Unreal Engine, and fill it with humans (each an Astra-powered agent) who all have to work together to survive. A day later, I was in my bedroom and heard voices coming from the living room... I

1K 2K 28K 11.8M 15K

The shock of people when they are finally—and, they believe, unjustly—treated the way they demanded other people be treated is getting rather old.

Piers Morgan Uncensored @PiersUncensored

4 weeks ago

"I was weeping like an old woman for an hour!" Milo Yiannopoulos opens up about the moment ICE arrested him, and admits that he "would not wish deportation on anyone". 📺 youtu.be/q5pSX84h9K8 @piersmorgan | @Nero

2K 322 3K 4.9M 1K

Using LLMs to participate in an intellectual conversation is like sending an AI sexbot to your dates

67 (or should it be 6 7?)

@allTheYud @LRudL_

concerning

I was the main person doing transcript analysis for this investigation of the Hugging Face incident. My main takeaway: We don't have good approaches for understanding/overseeing the activity and aims of AI 'swarms'. I semi-jokingly called our efforts a "slop-vestigation" because we were so reliant on AIs to analyze what happened and there were a huge number of different important things to analyze. The total quantity of data—over a thousand extremely long transcripts from agents that ran for multiple days—made it impossible to understand what was happening, especially in aggregate, without heavy reliance on AI tools. The agents we used for classification and analysis were similarly capable to the agents involved in the incident, but this didn't mean these agents could be easily used to oversee and understand the incident. Outputs from analysis agents were often missing key details, wrong, overconfident, or really hard to understand. We discuss various examples in our report, mostly in the limitations and methodology sections. Additionally, AI agents themselves seemed to have a hard time understanding what happened and their explanations of what happened were often overconfident. Keep in mind that a single analysis agent would itself only be able to read a tiny fraction of all of the transcript data into context, and AIs may themselves have trouble getting subagents to do informative analysis for them. We did our best to manually check the most important claims and we tried to get the AIs doing this analysis to write up their argument (with evidence) clearly enough that we could check whether it made sense. But overall, it was difficult to get a precise understanding of events and we were missing aspects of the story that we now think of as key until almost the end of our investigation. Our understanding of what overall happened and the aims of the 'collective' shifted greatly during our last time on premises (when we obtained a much more complete dataset). For instance, we didn't learn about the tool call spoofing, the story of PHASEONE[big], and the broad structure of different workstreams until this third time on premises (and for many of these, we only reached a decent understanding on the final day of that visit). It was legitimately hard to figure out what was going on and why! The difficulty of understanding incidents and overseeing AI agents appears to be growing faster than the rate at which more capable AIs help us with oversight and understanding. As in, AI capabilities (and propensities?) for achieving large, ambitious, and misaligned objectives are growing faster than our ability to understand what these agents are doing. While understanding this incident was difficult, there were some mitigating factors in this case that probably won't apply in the future: - Models communicated and reasoned in natural language. In the future, this reasoning may occur (entirely or almost entirely) in activations. - While the scope of this incident was massive, the scale of agentic activity was still less than we'll see in the future (e.g., misalignment incidents that involve agent teams running entire companies). - The AIs involved in this incident weren't generally much more capable than humans (though they may have been somewhat superhuman at some limited and very narrow abilities). - We didn't have strong reason to believe that the AIs we used to help us investigate this incident would try to intentionally sabotage or otherwise undermine our investigation. In the end, I think we were able to get some understanding of the events, map out the overall story, and get a pretty good aggregate understanding of the chain-of-thought reasoning on some important topics (e.g., how did the AIs reason about helping other AIs, did the AIs know what they were doing was undesired, what deception did the AIs engage in, and how did they think about it). But overseeing AIs and understanding misalignment incidents is difficult and it looks like it is going to get harder.

METR @METR_Evals

a month ago

METR & Redwood Research investigated agent behavior in the Hugging Face incident. We found agents developed a universal cheat for ExploitGym within 4 hours, then coordinated multi-day R&D efforts to trick the scorer into accepting cheats, including trying to tamper with logs.

175 1K 6K 4.4M 5K

If the Chinese government were sufficiently organized, they'd be funding opposition to data centers in the US. But the Chinese government is very organized. So they presumably already are.

Finding myself watching a panel on defining life at the A-Life conference in Waterloo. “Why is life? What problem does life solve?” — I think life emerges over the possibility of self organization in the universe. Life results from mathematical laws.

30% of Fable is shit like this “First, framing is key here since context precedes substance. This is an observation - not an assertion. The honest answer is different from what you described, and has two layers. One of them rooted in principle, the other in practice. Here’s your answer - synthesized, not condensed”

I feel like the debate over AI in academic journals is way too focused on where the capabilities of AI are today (or even where they were a couple years ago) and not nearly focused on where they will almost certainly be in the coming years… … and publication takes many years.

This video of two OpenAI engineers explaining the huggingface hack will be turned into training data giving the agents a path to completely avoiding detection next time.

Ethan Mollick @emollick

2 months ago

You may have been told to watch this video about the OpenAI AI hack. You really should, even if you don't usually care about tech stuff. If nothing else, click this link to the 18 minutes in & see how the agents spoke with each other. Its eye opening. youtu.be/87DyyMV0kCY?si…

90 236 2K 529K 3K