Ben Goertzel Revisits Omohundro's Basic AI Drives, Identifies Six Hidden Assumptions
bengoertzel · x · 2026-10-13
Ben Goertzel launches a five-part blog series revisiting Steve Omohundro's 2008 argument that advanced AIs will inevitably self-improve, self-protect, acquire resources, and resist shutdown—on which much AI-risk thinking rests. Part 1 writes the argument out step by step, identifies its six underlying assumptions, and asks which minds satisfy them. His finding: for minds meeting those assumptions the drives are an attractor, but for differently built minds the argument fails and other 'basic AI drives' emerge.
More from AGI Musings
- What's the steelman for AI shopping assistants actually picking the best product? — Afinetheorem · 2026-10-13
- "More economists should read Player Piano": Vonnegut's automation classic resurfaces in AI debate — soumitrashukla9 · 2026-10-13
- When customers become agents, how will they discover and buy your product? — jia_seed · 2026-10-13
- Jason: 99% of enterprises don't need esoteric math prizes, open source models will win — markjeffrey · 2026-10-13
- 1 in 5 words in Australian politicians' speeches are LLM-drafted, top in the world — JeremyNguyenPhD · 2026-10-13
- When advisors co-author without contributing: academia's authorship ethics debate — skoularidou · 2026-10-13