Our blog
Latest from Apollo

16 November 2026
A Pilot Towards Embedded Evaluations
Our pilot with OpenAI: six evaluators, four months across training, evals and internal deployment, with public reports and meta-transparency on our limits.
Read more
01 October 2026
Towards embedded evaluations for scheming propensities
Embedded evaluations for frontier AI scheming: four claims every safety case must make, and the persistent, deep access embedded evaluators need to verify them.
Read more
30 September 2026
On Testifying on Misaligned AI in the U.S. Senate
Apollo Research CEO Marius Hobbhahn testifies before the U.S. Senate on AI scheming: why our tools for detecting misaligned AI are degrading, and four steps to act before models learn to hide deception.
Read more