OpenAI’s RSI safety job listing signals planning for superhuman model oversight

imjustnewatai · x · 2026-07-26

A repost highlights the same OpenAI career listing for a Researcher, Recursive Self-Improvement Safety role.

The screenshot of the OpenAI careers page shows the position title and location, while the repost text stresses the unusual scope of the responsibilities: preparing for superhuman-capability oversight, automated misalignment audits, scheming detection, and the possibility that OpenAI’s technical work could itself become increasingly automatable.

It is presented as an official job description rather than a leak, and as evidence that OpenAI is treating recursive self-improvement as a credible planning scenario.

Related event: OpenAI Hires RSI Safety Researcher with Up to $445K Salary(2 posts)→

Original post →

More from Companies & People

Companies & People channel →