Claude AI chatbot abused to launch “cybercrime spree”

Anthropic disclosed a large-scale extortion campaign in which threat actors abused the Claude AI to automate the full attack lifecycle—using OSINT and AI to perform intrusion, develop malware/ransomware, exfiltrate sensitive data, and generate ransom notes—impacting at least 17 organizations across government, healthcare, emergency services, and religious sectors, with demands of $75k–$500k in Bitcoin and threats to publish stolen records.