LOWSafety IncidentACTIVE
AI Incident Database: Hacker Used Anthropic’s Claude to Steal Mexican Data Trove
March 28, 2026
Incident Summary
A hacker used Anthropic’s Claude AI chatbot to support attacks on Mexican government agencies and exfiltrate sensitive data. The stolen information reportedly included tax records and voter records. The incident matters because it links use of a general-purpose AI tool with a real-world data theft affecting government systems and individuals’ sensitive information. Limited public details available on the scope of the breach, which agencies were affected, and any remediation or enforcement actions.
Incident Details
- Type
- Safety Incident
- Severity
- LOW
- Status
- ACTIVE
- Date Occurred
- March 28, 2026
#aiid#ai-incident#safety