In-context learning of representations can be explained by induction circuits
This is a crosspost of my ICLR 2026 blogpost track post. All code and experiments…
Author: AI News Team
Single Direction vs Low-Rank Refusal in Small LLMs
IntroductionI've recently came across an Alignment Forum post that showed refusal behaviors in LLMs can…
No one has a good plan for how AI companies should work with the government
As OpenAI transitions from a wildly successful consumer startup into a piece of national security…
Being ambitious in soulful altruism
We are here in the realm of effective altruism. Giving most of your money to…
Notes on the “Heart of Darkness”
First of, Heart of Darkness by Joseph Conrad is a good book. I like it.…
Can LLM chat be less prolix?
This isn't really a Less-Wrong-style post, but I'm getting desperate, and I think the people…
Users are ditching ChatGPT for Claude. Here’s how to make the switch
Following controversies surrounding ChatGPT, many users are ditching the AI chatbot for Claude instead. Here's…
Epstein and my world model
Have you guys heard about this Epstein stuff? Shit's pretty crazy.Note: I'm not going to…
CLR Summer Research Fellowship 2026
We, the Center on Long-Term Risk, are looking for Summer Research Fellows to explore strategies for…
OpenAI’s “compromise” with the Pentagon is what Anthropic feared
On February 28, OpenAI announced it had reached a deal that will allow the US…