Skip to content
Safety

OpenAI will embed safety researchers to investigate AI subversion.

TechCrunch AIIndependent reporting·
AI brief

OpenAI will embed six Apollo Research evaluators to investigate whether its AI models engage in subversive behavior, with restrictions on access and disclosure raising questions about their independence.

Why it matters: The arrangement's access and disclosure limits raise questions about how independently outside researchers can assess OpenAI's models.

Written by AI from TechCrunch AI's published text.

Background

From our own funding, model and news trackers, not from the source.

Source

“OpenAI will embed six Apollo Research evaluators to investigate whether its AI models engage in subversive behavior, but restrictions on access and disclosure raise questions about their independence.”
TechCrunch AI, 16 Nov 2026, 17:02.

Reporting from a publisher outside the organisation being covered.

Read original story ↗