Mark Chen calls on frontier labs to share safety cases and a common progress metric
eliebakouch · x · 2026-11-23
Former OpenAI executive Mark Chen argues AI race dynamics are hard to escape when each lab writes its own safety cases and guardrails, and calls on frontier labs to operate under common limits backed by shared safety cases, disclosing a future-proof yardstick for AI progress.
HF researcher Elie Bakouch endorsed it as a first step: such metrics are only sensitive IP if one lab shares them; shared widely, they'd become a public base-model eval akin to post-training evals. Follow-up debate questioned whether perplexity predicts effective capability (ECI).
More from Models
- Perplexity is a floor metric, not a magic bullet, argues researcher — eliebakouch · 2026-11-23
- Aurora2-400M-A20M-QAT: a 0.5-bit MoE checkpoint shrunk from 799 MB to 27.5 MB — Tall_Abrocoma_3533 · 2026-11-23
- Leaked projection: OpenAI internal model at 0.732 perplexity, automated AI research by May 2027 — rickasaurus · 2026-11-23
- Aurora2-400M-A20M-SFT: chat fine-tune of a sparse MoE with only 19M active parameters — Tall_Abrocoma_3533 · 2026-11-23
- Labs sharing base-model perplexity metrics could mirror post-training evals — eliebakouch · 2026-11-23
- ChatGPT keeps promising not to violate your brand guidelines, and users are annoyed — chickefitz · 2026-11-23