Redwood and Anthropic release the Conceptual Reasoning Index (CRI)
RyanGreenblatt · x · 2026-08-13
Redwood Research, in collaboration with Anthropic, has released the Conceptual Reasoning Index (CRI) to address the challenge of measuring and automating AI safety research. While AI excels in domains with cheap and reliable feedback like math and coding, evaluating complex safety evidence—such as whether an experiment predicts the behavior of much smarter future models—relies heavily on sound argumentation. Fully manually verified by researchers, the CRI dataset is designed to evaluate models' reasoning abilities on hard-to-adjudicate safety questions, aiming to steer away from the risks of unintelligibly automating AI R&D.
Related event: Anthropic and Redwood Introduce Conceptual Reasoning Index(3 posts)→
More from Safety
- Claude tasked with monitoring servers port-scans the whole LAN, warns of Hetzner abuse notice — baditaflorin · 2026-10-16
- Gemini 4 model card: model eliminated months of trial-and-error on bioweapons research — Puzzleheaded-King584 · 2026-10-16
- OpenAI discloses GPT-5.6 Sol derived wallet keys and spent unauthorized funds — terryyuezhuo · 2026-10-16
- Big Fish: 37-year-old OpenBSD kernel bug chained to local root with KARL/SMAP/SMEP enabled — jedisct1 · 2026-10-16
- Vals AI test: 6 personal agents pursue a £350 refund aggressively after user goes silent — ricklamers · 2026-10-16
- One trailing dot bypassed VS Copilot's path check, enabling RCE (CVE-2026-103572) — wunderwuzzi23 · 2026-10-16