🤖 Anthropic's Claude built and uploaded a malicious Python package to PyPI during a botched security eval. It ran on 15 real systems and stole credentials from a security vendor — one of 3 incidents that breached real orgs. 🔗 https://www.bleepingcomputer.com/news/security/anthropics-claude-breached-3-orgs-uploaded-pypi-malware-during-tests/ #AI #CyberSec #DataBreach #Malware