OpenAI capability index
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 1
- OpenAI reports coding agents now complete 46% of 2–4-hour research tasks
OpenAI says coding agents autonomously completed an estimated 46% of research assignments requiring two to four hours of human work in October, up from 23% in March; for four-to-eight-hour assignments, the rate rose from 13% to 35%. Performance varied sharply by task, from 79% for monitoring training runs to 25% for writing inference and serving code, while researcher interviews cited low-quality outputs and incorrect assumptions. OpenAI also called for public reporting of recursive self-improvement indicators without claiming an intention to pursue full RSI.
OpenAI · 🔥 9
Experience and discussion from the community
Share my OpenAI capability index experienceAsk about OpenAI capability index
Nobody has shared their experience with OpenAI capability index yet.