Navigation

Co-authored Risks from Learned Optimization (2019) introducing mesa-optimization and deceptive alignment; led Sleeper Agents and Alignment Faking research at Anthropic; 3,400+ citations

Core Data

Entity	Evan Hubinger
Property	Notable For
Formatted Value	Co-authored Risks from Learned Optimization (2019) introducing mesa-optimization and deceptive alignment; led Sleeper Agents and Alignment Faking research at Anthropic; 3,400+ citations
Raw Value	Co-authored Risks from Learned Optimization (2019) introducing mesa-optimization and deceptive alignment; led Sleeper Agents and Alignment Faking research at Anthropic; 3,400+ citations
Value Type	text
Unit	—
As Of	Mar 2026
Valid End	—
Expired?	No

Fact ID	f_eH6tN2pQ5v
Subject ID	XnqqKHiNmw
Property ID	notable-for
Derived From	—
YAML File	packages/kb/data/things/XnqqKHiNmw.yaml