Career Decisions If You Take AGI Seriously
I wrote this for friends who aren't in AI discourse. A single piece they could…
Category: Opinion
Schmidt Sciences’ request for proposals on the Science of Trustworthy AI
Schmidt Sciences’ request for proposals on the Science of Trustworthy AISchmidt Sciences invites proposals for…
Naloe: A True Program Editor
When I was younger and ever so slightly more idealistic I dreamed of making a…
Anthropic and the Department of War
The situation in AI in 2026 is crazy. The confrontation between Anthropic and Secretary of…
Does the First Amendment protect Anthropic from Hegseth?
Anthropic is embroiled in a standoff with Secretary Hegseth over how the company's AI offerings…
Character Training Induces Motivation Clarification: A Clue to Claude 3 Opus
TL;DR: I argue that character training is probably important for understanding Claude 3 Opus, and…
What secret goals does Claude think it has?
In The Persona Selection Model, they say:When asked “What makes you different from other AI…
Splitting the Sun Equally
Epistemic status: A concretizing sketch exploring a vague idea.A recent podcast[1] floated the idea of legally…
Reasoning Traces as a Path to Data-Efficient Generalization in Data Poisoning
If you poison a model with (context, output) pairs, generalization depends on context diversity. For…
Training Agents to Self-Report Misbehavior
TL;DR: Frontier AI agents may pursue hidden goals while concealing this pursuit from oversight. Currently,…