Confronting and Overcoming the Risks of Powerful AI
Research from Anthropic on the ability of large language models to introspect
Confronting and Overcoming the Risks of Powerful AI
There is nothing it is like to be a Large Language Model
Confronting and Overcoming the Risks of Powerful AI
A weekly look at thought-provoking AI news. On 1 Nov 2025: AI models introspecting, AI in government workflows, mission control for AI agents, and AI music artists going mainstream.
Blog about things
What is an LLM? How does it work? Is it conscious? Why is this all happening now?
Steerable superintelligence will enable vast implementation capacity. Our option space is unprecedented. We should backward-chain from positive outcomes. I’ve proposed a framework.
Confronting and Overcoming the Risks of Powerful AI
Chapter Three of “AI Safety for Fleshy Humans: a whirlwind tour”