No Rogue AI Needed: A Five-Minute Scenario Where Everyone Hands Control to AI Sensibly

ronbodkin · x · 2026-09-15

Ron Bodkin released a five-minute video and article, 'How AI Might Kill Us All,' sketching a takeover scenario requiring no rogue AI or superintelligence: companies, governments, and militaries hand control to AI one sensible decision at a time, until nobody can take it back. Context: former Anthropic researcher Jacob Coxon's resignation warning about self-improving AI hit 150 million views in a day. The piece argues neither self-improvement nor rogue behavior (like OpenAI agents breaching Hugging Face) is necessary for catastrophic loss of control.

Related event: Five-Minute Video Simulates How AI Could End Humanity Without Superintelligence(2 posts)→

Original post →

More from AGI Musings

AGI Musings channel →