We study alignment audits—systematic investigations into whether an AI is pursuing hidden objectives—by training a model with a hidden misaligned... more

The Most Forbidden Technique is training an AI using interpretability techniques. An AI produces a final output [X] via... more

You learn the rules as soon as you’re old enough to speak. Don’t talk to jabberjays. You recite them as... more

Exciting Update: OpenAI has released this blog post and paper which makes me very happy. It's basically the first steps... more

LLM-based coding-assistance tools have been out for ~2 years now. Many developers have been reporting that this is dramatically increasing... more

Background: After the release of Claude 3.7 Sonnet,[1] an Anthropic employee started livestreaming Claude trying to play through Pokémon Red.... more

Note: an audio narration is not available for this article. Please see the original text. The original... more

In a recent post, Cole Wyeth makes a bold claim: . . . there is one crucial test (yes... more

This isn't really a "timeline", as such – I don't know the timings – but this is my current, fairly... more

This is a critique of How to Make Superbabies on LessWrong. Disclaimer: I am not a geneticist[1], and I've... more