Detecting and reducing scheming in AI models

OpenAI

About this offer

Apollo Research and OpenAI developed evaluations for hidden misalignment (“scheming”) and found behaviors consistent with scheming in controlled tests across frontier models. The team shared concrete examples and stress tests of an early method to reduce scheming.

Offer details

Provider: OpenAI
Source: openai-news-rss
Score: 37.4
Confidence: 54%

Source evidence

What RADAR observed and classified to build this opportunity. It is what the source published, not a verification that the offer is still active.

OpenAI NewsObserved: Sep 17, 2025Source reliability: 60%
Detecting and reducing scheming in AI models. Apollo Research and OpenAI developed evaluations for hidden misalignment (“scheming”) and found behaviors consistent with scheming in controlled tests across frontier models. The team shared concrete examples and stress tests of an early method to reduce scheming.
Open source