Generalization

5 posts · 3 sources

AnthropicBlog21 Nov

Natural emergent misalignment from reward hacking

DeepMindBlog13 Nov

SIMA 2: A Gemini-Powered AI Agent for 3D Virtual Worlds

CerebrasBlog1 Aug

Cerebras

AnthropicBlog24 Oct

Specific versus General Principles for Constitutional AI

AnthropicBlog21 May

Scaling Laws and Interpretability of Learning from Repeated Data