Apollo Research CEO Marius Hobbhahn testifies before US Senate on AI scheming
Apollo Research says tools for detecting misaligned AI are degrading and points to four steps to act before models learn to hide deception.
TLDR
Marius Hobbhahn said he testified before the US Senate on rogue AI. Apollo Research describes the testimony as covering AI scheming, why tools for detecting misaligned AI are degrading, and four steps to take before models learn to hide deception.
Combined views
—
1 Source, first seen ago
