Jacob Steinhardt maps concrete pathways to losing control of advanced AI
Jacob Steinhardt argued that opaque, agent-like, rapidly changing AI could amplify cyberattacks, unemployment, mis-optimization, and loss of human sovereignty, while outlining value learning, verification, transparency, and strategic planning as near-term research priorities.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why it moved the index
The essay contributes a specific early synthesis of control difficulty: open-ended agency, rapid capability change, opacity, weakly supervised value learning, cyber misuse, labor disruption, and diminished human sovereignty. Magnitude is moderate and confidence is bounded because these are reasoned mechanisms and forecasts rather than a completed loss-of-control event.
Assessment history
-
R1
Toward 40 · confidence 72
Initial historical inclusion from a fully dated first-person essay with distinct control-risk reasoning.
17 Aug 2026