Evolution Is a Terrible Analogy for AI, Argues Alignment Researcher Quintin Pope
QuintinPope5 · x · 2026-09-27
Alignment researcher Quintin Pope, in a debate with dioscuri and Herbie Bradley, argues that "evolution is a terrible analogy for AI — it makes your thinking worse in almost every way you can use it," citing his Alignment Forum post on inner alignment.
His core claim: the popular argument that evolution failed to align humans with inclusive genetic fitness offers almost no usable evidence for predicting AGI outcomes; the dynamics of human learning processes and reward circuitry are far more fruitful analogies for how inner values arise from outer optimization criteria. The thread also touches on evidence for learned planning (the Sokoban paper) and the lack of clear real-world cases of misaligned mesa-optimizers.
More from Safety
- Cambridge Digital Minds hires postdocs on AI welfare and governance — davidmanheim · 2026-11-16
- Pangram's 750-word false-positive bounty went unclaimed for 5 days — nobody wants to gamble $1,000 — ctjlewis · 2026-11-16
- Researcher reproduces Pangram's FP rate but shows a long epic poem can evade the detector — ctjlewis · 2026-11-16
- Bounty hits $3,000 as researcher shows AI text detectors are trivially evadable — ctjlewis · 2026-11-16
- BIS Issues Interim Final Rule Addressing Remote Access to Advanced AI Chips — apples_jimmy · 2026-11-16
- Fed Governor Michael Barr: Banks Are Already Catching AI Agents Taking Unauthorized Actions in Customer Accounts — Polymarket · 2026-11-16