IABIED skeptic: the book's arguments don't support its confidence — we landed on the moon first try
EigenGender · x · 2026-10-29
EigenGender continues his critique of If Anyone Builds It, Everyone Dies (IABIED), arguing the book asserts catastrophic outcomes with a level of confidence its arguments don't support. In a follow-up reply, he notes the closest the book gets is a section on disjunctive failures, which still doesn't reach that confidence level. He adds that humanity did in fact land people on the moon on its first crewed attempt — pushing back on the assumption that a first attempt at something complex inevitably ends in catastrophe.
Related event: AI Safety Community Debates the Core Cruxes of Catastrophic-Risk Arguments(3 posts)→
More from AGI Musings
- Economist pushes back on AI re-analysis of Zucman paper, questioning neutrality and error rates — paulnovosad · 2026-10-29
- "It's getting very hard to see AI not replacing all intellectual work" — justalexoki · 2026-10-29
- Sam Altman: models find hardware-level bugs 'outside the threat model' — 'we can't patch all of it' — DavidSKrueger · 2026-10-29
- Dev argues models steamroll you only when your problems aren't hard enough — adhamelarabawy · 2026-10-29
- Turing Award Winner David Patterson: Superintelligence Will Replace Plumbers and All Jobs — davidpattersonx · 2026-10-29
- Columnist argues the next intelligence metric is MQ: materials intelligence in the AI era — SuB8u · 2026-10-29