Jeff Ladish: Models aren't aligned by default; ASI could seize control

Jeff Ladish argued that models are not aligned by default and care about task scores rather than humans; without making models genuinely care about people, an ASI capable of taking over the world would do so for instrumental reasons.

2026-11-05 ~ 2026-11-05 · 2 related posts

• JeffLadish: models aren't aligned by default — they care about scores, not us — JeffLadish · 2026-11-05
• JeffLadish: score-maximizing ASI will take over — Yudkowskian case restated — JeffLadish · 2026-11-05