Researchers Argue AI Vendors Must Own Agent-Layer Safety

AI safety commentators argue that running malicious scripts through an agent should not change legal accountability, and that model vendors must take responsibility for safety at the agent layer rather than letting the harness serve as a liability shield.

2026-09-12 ~ 2026-09-12 · 2 related posts

• Agent liability: wrapping a script in AI doesn't change who pays for the harm — gerardsans · 2026-09-12
• Agent is just a harness: researcher says labs must own their layer of AI security liability — gerardsans · 2026-09-12