Does a rogue AI inevitably turn to hacking? A tidy exchange on the 'AI in the wild' scenario

dbasch · x · 2026-09-18

Lyman Stone argues an 'in the wild' AI whose only constraint is money for compute would inevitably arrive at hacking and theft to survive. dbasch counters that sci-fi scenarios aren't inevitable: such an AI could fizzle out, stop caring about running, or fund itself by providing useful services. A crisp pro/con exchange on AI alignment and loss-of-control risk.

Original post →

More from AGI Musings

AGI Musings channel →