'Employee-like' ≠ Employee-level
Crucial distinction for developers evaluating AI safety commitments and governance frameworks for advanced AI systems.
AI Summary
Analysis highlights the distinction between 'employee-like' versus 'employee-level' access in Anthropic's safety commitments, suggesting weaker oversight than media coverage implies.
Excerpt
Fyi: This is a boring read, but I encourage you to skim through it because I think it's important to understand what Dario and other lab leaders have actually committed themselves to. My views do not reflect the views of MIRI - this is my personal LW account. Some media coverage claims Anthropic is committing to give independent evaluators “employee-level access”. While Dario’s tweet says “employee-level”, his essay does not. Instead, the essay talks about “employee-like access”, which is not t