arXiv:2610.11093 · v3

Risk Ceilings and Development Deadlines: Pacing AI under Uncertain Safety Productivity

首次提交 2026-10-08 10:10

版本更新 2026-11-06 01:33

雷达观测 2026-11-06 12:09

摘要

Can a regulator promise both a risk ceiling and a development deadline when safety productivity is unknown? A ceiling below the final model's unprotected hazard requires a minimum stock of safety knowledge, so both promises hold only if weak research can be ruled out. Learning first and then replaying development, with spare compute in safety, comes close to that minimum. In a calibration with only state risk, constant safety yield, and full knowledge transfer, it needs only 3.3 percent more productivity than the necessary bound. Customer services pay for the guarantee. Rules that fix compute allocation fix dates, not risk.

作者备注：v2: full-length version, with the supplemental appendix included as Appendices C-J; v1 was a short version of the same paper