DeepMind researcher: human inconsistency makes perfect AI alignment impossible in principle
jachiam0 · x · 2026-10-16
DeepMind researcher Janus (jachiam0) argues alignment is fundamentally hard because humans don't truly know what they want and can only imperfectly express it. Even if AI could read minds and chart a path to a CEV, we would reject the gilded cage. Our inconsistencies both preclude perfect alignment and underlie much of what is beautiful in human experience, so enforcing conformity would destroy what we're trying to protect. We'll have to muddle through, under high uncertainty, in a universe stratified by an intelligence layer far smarter than us.
More from AGI Musings
- When Simple Rules Beat AI: A Frozen Ride-Hailing State Machine at Dulles — shashib · 2026-10-16
- PhD advisor buys every student a Codex Pro subscription: execution is getting cheap — DJiafei · 2026-10-16
- Could a baby ASI survive without humans using today's robotics? Viral thread argues it's a timing problem — NathanpmYoung · 2026-10-16
- The 'infrastructurization of everything': only compute, storage and inference will be left to sell — AccBalanced · 2026-10-16
- Dwarkesh Patel in NYT: AI will do all work in the economy by 2030 — Traditional-Chip8339 · 2026-10-16
- "Every week during the singularity is just utterly insane," dev says on AI's pace — tszzl · 2026-10-16