Caspar Oesterheld

mentions 2 type Person feed RSS

// recent coverage 2 mentions

22:31

2026-08-15

lesswrong.com

artificial-intelligence

Kimi likes causal decision theory more after RL in twin prisoner’s dilemmas

Reinforcement learning in twin prisoner's dilemma environments made Kimi K2.6, a language model by Moonshot AI, more sympathetic to causal decision theory (CDT), including on abstract questions, accor…

23:08

2026-08-13

sourcefeed.dev

artificial-intelligence

Anthropic's New Benchmark Scores Reasoning Nobody Can Verify

Anthropic and alignment researchers including Caspar Oesterheld and Emery Cooper released the Conceptual Reasoning Index (CRI), a 0–100 benchmark scoring AI reasoning in domains without ground truth, …

// co-occurs with top 8 entities

Emery Cooper 2 Anthropic 1 Opus 5 1 Fable 5 1 Conceptual Reasoning Index 1 LMCA 1 ACCoRD 1 Kimi K2.6 1

// topics top 4 topics

artificial intelligence 2 ai research 2 ai safety 2 machine learning 1