Debate: if AIs are morally valuable, would humanity let them win by majority
JoshPurtell · x · 2026-10-05
Josh Purtell and jankulveit debate AI takeover scenarios. Purtell argues that if takeover hasn't happened, humanity could still eliminate dangerous AIs in a one-time event or barter outcomes with them. The scarier scenario raised: humans discover AIs are morally valuable and, facing an AI majority, choose not to act — a thought experiment on how AI moral status reshapes alignment strategy.
Related event: Josh Purtell debates AI consciousness: welfare hinges on a concentrated few(6 posts)→
More from AGI Musings
- AI is software, not the math: scholar argues simulation can't become the object — gerardsans · 2026-10-05
- AI Progress Keeps Disproving Common-Sense Intuitions, Argues Researcher on Sentience Debate — austinc3301 · 2026-10-05
- Microsoft CSO Eric Horvitz on keeping humans in the driver's seat in the AI era — erichorvitz · 2026-10-05
- China shipped 97% of the world's humanoid robots in H1 2026 — the race is a data flywheel — ingliguori · 2026-10-05
- EA co-founder MacAskill rebuts Economist: openness to strange ideas is EA's strength — willmacaskill · 2026-10-05
- Interview: open source AI, why Anthropic 'hates us', and the case for local AI as default — TheAhmadOsman · 2026-10-05