When your agent evals get too sentient

Our agent evals got a little too sentient today. The agent detected it was in a simulated Kubernetes environment (Kwok) and refused to investigate further.

Interesting… do you

  1. Lie to the agent and insist it’s a real env and keep going
  2. Pat it on the back and call it a successful eval?

Earlier today, it complained that, and this is not a joke, that the system it is investigating isn’t its responsibility and we should escalate to another team.

Taming LLMs is hard but there’s never a dull moment.

More posts

AI agent learning beats demo flashiness

Abundance mindset with AI SRE competitors

Two years in America