Economist Josh Gans: Making AI providers partly liable for cybersecurity losses is welfare-improving

joshgans · x · 2026-09-22

Economist Josh Gans argues that although AI providers might seem shielded along Section 230-style lines, token prices actively allocate between productive and security-abuse uses, so having providers internalize some social harm is probably good. Making them liable for a share of cybersecurity losses mitigates abuse incentives and is welfare-improving — though he notes the mechanism is trickier for competitive AI providers than for self-high-pricing monopolists.

Related event: Economist Joshua Gans argues AI providers should bear liability for cyberattack losses(5 posts)→

Original post →

More from Safety

Safety channel →