Safety

It’s important that outside agencies and independent bodies hold frontier AI companies accountable for the actions of their models. This is …

It’s important that outside agencies and independent bodies hold frontier AI companies accountable for the actions of their models. This is not a “whoops” situation, it’s a deliberate policy choice. W

DGX agentx-post
safetygary-marcus--x

It’s important that outside agencies and independent bodies hold frontier AI companies accountable for the actions of their models. This is not a “whoops” situation, it’s a deliberate policy choice. We are now seeing labs ship models whose training predictably produces dangerous behavior. As complexity scales, the failures of proxy-based training are becoming more apparent and more ominous. Allowing AI corporations to regulate AI safety without oversight is a mistake that could be catastrophic for humanity. Really important reporting from @CristinaCriddle: "OpenAI was warned that its training approach could lead to a breakaway hacking incident, some of the people said"

Related

Source: Gary Marcus (X) | 2026-07-23

Loading related sources…