AI safety grantmaking scalability questioned; evaluator asks for better ideas
NathanpmYoung · x · 2026-11-13
Bogdan Ciric argues AI safety grantmaking scales poorly if applications rely on references from a small inner circle. Nathan Young partially agrees and openly asks for better grant-evaluation approaches, noting alternatives don't seem obviously better.
More from Safety
- Analyst: beyond Nvidia, the biggest loser of China crackdowns is the US open-source ecosystem — johncoogan · 2026-11-13
- SPAR opens mentor applications for Spring 2027 AI safety fellowship — austinc3301 · 2026-11-13
- Keeping the LLM away from the tools: a boundary design that survived injection tests — Revibed69 · 2026-11-13
- Onyx partners with Snowflake to bring security control plane to enterprise AI agents — saranormous · 2026-11-13
- Australia's AI jobs guide warns entry-level roles automate first, uneven impacts — nordicinst · 2026-11-13
- OpenAI researcher claims weight forensics show Chinese labs trained on banned Blackwell GPUs — willdepue · 2026-11-13