#AI-Safety

Claude AI Exploited to Automate Mexican Government Network Breach

Unknown actors bypassed Anthropic's Claude safety filters to automate vulnerability discovery and data exfiltration against Mexican government systems.