Anthropic Faces Multiple Human-Caused Incidents in March 2026

Summary

TechCrunch reports that Anthropic experienced at least two notable incidents caused by human error within a single week in late March 2026. The article, authored by Connie Loizos, highlights the irony that Anthropic — widely recognized for its detailed AI risk research, top-tier safety researchers, and vocal stance on the responsibilities of building powerful AI — has been tripped up by its own people rather than its technology. While the full details of the incidents are not available from the