Early Anthropic hire and ex-METR COO launch startup to rein in rogue AI agents

ThereWas · reddit · 2026-09-15

Per TechCrunch, an early Anthropic hire and the former COO of model-evaluation org METR have founded a startup focused on reining in rogue AI agents.

METR is known for evaluating dangerous capabilities of frontier models; the founders are turning that evaluation expertise into a product play for agent safety — a sign that agent control is moving from research topic to commercial product.

Original post →

More from coding & agent

coding & agent channel →