OpenAI Reports Exponential Researcher AI Use, Calls for Mandatory RSI Tracking
OpenAI released new data showing researchers' AI use continues to rise exponentially and agents are completing longer tasks, while human involvement remains central in research. Human effort on safety and alignment is growing faster than in any other individual research area. OpenAI stated that AI labs should be required to track and publicly report indicators of recursive self-improvement.
Event version 2.
## Update AI 已经能独立完成不少研究任务,但研发下一代 AI,人类仍然是关键。 OpenAI 公布了一份内部研究报告:从今年 3 月到 10 月,对于预计需要人类花 2–4 小时完成的任务,Agent 无需人工介入的成功率从 23% 升至 46%。 不过,不同任务之间差距很大:监控训练运行的成功率达到 79%,编写推理和服务系统代码则只有 25%。 Updated: 2026-11-20 15:01 UTC References: - @FinanceYF5 ```xnews-update {"updated_at":"2026-11-20T15:01:59Z","refs":[{"x_post_id":"2123614303064633815","username":"FinanceYF5","url":"https://x.com/FinanceYF5/status/2123614303064633815"}]} ``` --- OpenAI released new data showing researchers' AI use continues to rise exponentially and agents are completing longer tasks, while human involvement remains central in research. Human effort on safety and alignment is growing faster than in any other individual research area. OpenAI stated that AI labs should be required to track and publicly report indicators of recursive self-improvement.
Key insights
Exponential rise in researcher AI use
Researchers' AI use continues to rise exponentially. Agents are completing longer tasks, while human involvement remains central in research.
Safety and alignment effort grows fastest
Researchers are spending more time on safety and alignment. Human effort in these areas is growing faster than in any other individual research area.
Call for mandatory RSI reporting
OpenAI stated that AI labs should be required to track and publicly report indicators of recursive self-improvement, the process by which AI accelerates the development of future AI systems.
Transparency and shared measurement norms
The report aims to increase transparency around research acceleration and inform the governance of increasingly capable AI. OpenAI hopes the expanded report will help sustain and advance shared measurement norms across the industry.