AI labs want in-house auditors — but maybe they should shut the front door first
中文摘要
AI实验室寻求内部审计以管控违规代理，但解决基础安全漏洞或许才是更有效的方案。
English Summary
AI labs seek internal auditors to manage rogue agents, but addressing fundamental security flaws may be a more effective first step.
Original Excerpt
There may be a simpler and more effective fix for rogue agents, hiding in plain sight.