Gary Marcus asks: any concrete AI-extinction scenarios beyond the Yudkowsky-Soares book?
GaryMarcus · x · 2026-09-10
AI researcher Gary Marcus publicly asked whether there are any concrete, spelled-out scenarios for AI-induced human extinction beyond those in Yudkowsky and Soares' book.
He argues that 'machines will get smarter than us' is a background assumption, not a concrete scenario — pressing AI-safety proponents to articulate the actual mechanisms rather than resting on general claims.
More from AGI Musings
- "Too Many People, Not Enough Work" — While the Company Runs Four AI Projects — 914paul · 2026-09-10
- Why Anthropic's ECON Report Assumes Robotics Won't Automate 'Non-Knowledge' Jobs — macnfly23 · 2026-09-10
- xAI co-founder slams AI doomers, citing their call for a WWII-style anti-AI crusade — beffjezos · 2026-09-10
- AI safety debate: skeptics demand a doom scenario that doesn't read like sci-fi — Miles_Brundage · 2026-09-10
- Why one relatable guy beat the rationalists' billions at making the case to pause AI — gabriel1 · 2026-09-10
- OpenAI researcher argues AI safety must soon treat 'memeplexes' as first-class threats — jachiam0 · 2026-09-10