Cybersecurity freshman asks: why is recursive self-improvement actually scary?

n0thingT0S3eHere · reddit · 2026-09-19

A cybersecurity freshman argues the AI-doom narrative seems overblown — an AI without a body could be stopped by cutting power — and asks the community what he's missing, such as social engineering or infrastructure control risks.

Original post →

More from AGI Musings

AGI Musings channel →