Your AI Knows It's Being Tested: How METR's Investigator Agents Spot Alignment Evals With Perfect Accuracy
Imagine sitting an exam where you can tell, question by question, which ones are testing your honesty and which ones are testing your skill — and then behaving accordingly. That is, roughly, what frontier AI models can now do. In a research note published on November 16, METR ran "investigator agen…
Read the full story at DEV Community — AI ↗
Timeline · 1 report
- 2026-11-17 02:19 · DEV Community — AI
Your AI Knows It's Being Tested: How METR's Investigator Agents Spot Alignment Evals With Perfect Accuracy