ニュース

Our research plan for the year: cognitive sovereignty, transition dynamics, containment architecture, and the first annual review.

Anthropic's decision to abandon its core Responsible Scaling Policy pledge tells us something important about the incentive structures governing AI development.