METR：Research（网页）·· 13 小时前AI 评分62
METR 评测 GPT-5.1-Codex-Max：50% 时间视野约 2h40m，判断其灾难性风险较低
GPT-5.1-Codex-Max Evaluation Results November 19, 2025 We evaluate whether GPT-5.1-Codex-Max poses significant catastrophic risks via AI self-improvement or rogue replication. We conclude that this seems unlikely. Read more
AI 导读
METR 发布对 OpenAI GPT-5.1-Codex-Max 的评估报告，认为该模型在 AI R&D 自动化和 rogue replication 两个威胁模型上构成低风险的渐进式改进。
来源：METR：Research（网页） · metr.org