ff

[Quantіan @quantian1]

You can do Fourier transforms in O(N log(N)^0.9999999999999) time. Fuck this gay chud universe

[יָאוּ @1YoGoHa]

אני כשהמטוס טיפה מתנדנד בגלל רוחות: *רועד מפחד בא לי למות*

ישראלים במטוס מדובאי כאשר המטוס צונח 15 אלף רגל בדקה בגלל ניסיון פיגוע:

[ff @F_gana]

This is the safety-sphere equivalent of Greg brocknan leaving OpenAI to join Anthropic

[Ryan Greenblatt @RyanGreenblatt]

I'm joining METR to work on more investigations like our Hugging Face report.

Currently, tons of even basic information about AI development that's highly relevant to catastrophic risk isn't public. I used to be more skeptical of the value of public info, but recent events have changed my mind.

Getting verified information about what's going on inside AI companies seems particularly urgent now. The limited public evidence we have seems consistent with the possibility that imminent recursive self-improvement could massively accelerate capabilities progress, which could then potentially yield extremely superhuman general capabilities within 6 months or a year. If this occurred, there would be a correspondingly large risk of worst-case outcomes. This uncertainty about extreme outcomes could be substantially resolved with more verified public information: we could either build more consensus about near-term risk or learn that such extreme outcomes are less likely in the near term.

Beyond AI capabilities and takeoff, the state of public evidence is also highly limited for alignment, security, control, and risk-relevant internal processes at AI companies. This makes it hard to determine exactly how well or poorly these key areas will go in the near future. (METR plans to focus, at least initially, on just capabilities/takeoff, alignment, and control; I hope other groups cover security, internal processes, and other important areas.)

While I'm no longer working at Redwood, I think the work they are doing is very important; I'm excited about Redwood's ongoing contributions to R&D on technical mitigations and better public interpretation of risk-relevant evidence.

[מלך התוכן 👑 @yifty_cent]

כולכם מזיינים את המוח על נז"א בלי שראיתם את הסרט. אם הייתם צופים הייתם יודעים שבסוף יובל אברהם אומר למצלמה: "אז מה למדנו היום? למדנו שגם אם הם עושים לפעמים טעויות, בסוף אין על החיילים שלנו. עם! ישראל! חי!"
ואז בקלוזינג קרדיטס יש ביצוע מיוחד ל"שבט אחים ואחיות"

[Jacob Coxon @hilbertspaess]

I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.

[shako @shakoistsLog]

tyler cowen: oh you’re a doomer? why don’t you take asset positions that only pay off if you’re dead.

[Evan Hubinger @EvanHub]

Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.

[Ilay (אלוף המדינה) Marom @ilay_marom]

פוב: אתה מהנדס נאZי אחרי המלחמה

[Ryan Greenblatt @RyanGreenblatt]

OpenAI's newest AI, Astra, is reported to use an 'opaque reasoning' architecture where more of the reasoning occurs in activations instead of natural language. This may be the single worst development for AI security/safety to date.

The details of Astra aren't publicly known, but some versions of these architectures could very strongly impair oversight and monitoring. Based on the article, it seems like the recurrent depth (amount of opaque reasoning) is limited. This would mean that the AI is still reliant on natural language chain-of-thought, though to a reduced extent, making the natural language reasoning (much?) less useful for monitoring/oversight.

My biggest concern is that a natural progression from here would involve scaling up the opaque reasoning to the point where the model reasons entirely or almost entirely in latent space. This would very likely destroy the usefulness of chain-of-thought for monitoring/oversight (especially when AIs are trying to avoid detection or there is optimization pressure against the chain-of-thought). I hope it isn't too late to avoid the most concerning architectures and that OpenAI will stop here.

Some concrete examples of why this is extremely dangerous (inspired by recent events):

- In our investigation of the OpenAI / Hugging Face incident, we were heavily reliant on chain-of-thought. If the AIs we were investigating had instead been reasoning in latent space, this would have greatly undermined our investigation. Getting a good understanding of the behavior of this many agents was tricky enough even with the use of chain-of-thought!
- The AIs in the Hugging Face incident were very interested in messing with their transcripts. As part of this, they engaged in tool call spoofing. If these AIs had been capable of doing extensive latent reasoning, it would have been much more feasible for them to use tool call spoofing to construct an elaborate false narrative about what was happening without this being clear in their reasoning or clear from prior tool calls. (They could use latent reasoning to devise and then execute a tool spoofing strategy that would be difficult to detect: steganographically encode the "real" tool call into each pretend tool call, then have their spoofing software execute this encoded real tool call and return an output which encodes the real result in some hard to spot way.)

At a more basic level, my sense from their public communications is that OpenAI plans on being extremely reliant on chain-of-thought monitoring for safety. This does not seem like a good development for that plan.

The public doesn't currently have enough information to assess exactly how bad the architecture of Astra is, though based on the discussion in the article, the direction it moves in is extremely concerning. OpenAI should release more information about Astra's architecture as well as how much this architecture degrades monitorability and increases the AI's ability to reason opaquely. Credible independent assessment (or verification of OpenAI's assessment) is also important.

It seems like we are now engaged in a race to the bottom on architectures that could be catastrophic for our ability to oversee/monitor AIs. (While simultaneously, AI companies believe they'll train extremely capable AIs within a few years.) It may not be too late for AI companies and employees at AI companies to take aggressive action to avoid the worst outcomes.

[Nathan Calvin @_NathanCalvin]

"we know OpenAI agents were trying to delete logs of their misbehavior, but we can't find any examples where they succeeded"

[ff @F_gana]

@TheAmazing_Tofu באנגליה זה הפוך.

[Corey Walker 🇺🇸 @CoreyWriting]

This entire video depicts Israelis smiling, dancing, and having fun while the narrator talks about them as if they are demons. The stark juxtaposition is notable, and reflective of this entire discourse in general. Israelis are generally happy and grateful people and the leftist activists are depressed, angry, delusional, and overall unpleasant.

I also notice how she claims Israelis are "colonizers" In Israel and also "colonizers" outside of Israel. She talks about them as if they should not exist. This implicitly eliminationist rhetoric subtly suggests Israelis should simply not live anywhere.

Also, notice how she does not distinguish between "liberal" Israelis and the military members. As I've said before, these leftists want all Israelis eliminated, irrespective of their actual ideology. Their goal is extermination, and no amount of deterrence to progressive talking points is going to save you.

This clip is a window into the mind of bloodthirsty third-worldists. The entire Israel-Palestine discussion cannot leave the discourse fast enough. People's brains are becoming rotted, and we are becoming much more like the global south.

[Corey Walker 🇺🇸 @CoreyWriting]

These people believe that Israelis don't belong in Israel and also don't belong anywhere else. Anywhere an Israeli goes, they are just a "colonizer."

I don't understand how any Jew can be an anti-Zionist. The third-world leftists are barely hiding what they plan on doing with you if Israel were ever to collapse. All the "social justice" rhetoric is just a placeholder until they actually get to implement their actual agenda.

[Dor.Y.S.M @DorDugy]

מה המונדיאל יכול ללמד רווקים ורווקות (??) על השקעות (????) בשיתוף תפוזינה (?!?!?!?!?????????)

[ff @F_gana]

(3/3)

Anyways, here is an interactive view
https://t.co/Yzb1BPBzX9

[ff @F_gana]

(1/3)

This year, San Francisco celebrated the 4th of July by launching fireworks from the Golden Gate Bridge! I wanted to see it, but I didn't know where to go, so I made a map to highlight all the "publicly accessible" places where the bridge is visible.

[ff @F_gana]

(2/3)

I found the map to be cool and it revealed some unexpected things. For example, it turns out that you can see the bridge all the way from a hill in Pacifica!

View from the hill in pacific (my rendering on the left, real view on the right)

[big content guy @bigcontentguy]

the english fascinate me with these chants. how does everyone know what the chant is gonna be me and spread the information so fast. we can barely do the wave

[ff @F_gana]

@BjarturTomas @koreindian done