Anthropic launches Claude Security in public beta for enterprise users
Updated
Updated · ZDNet · Apr 30
Anthropic launches Claude Security in public beta for enterprise users
13 articles · Updated · ZDNet · Apr 30
The tool uses Claude Opus 4.7 to scan full repositories or directories, generate targeted patches, and provide confidence-rated findings, with Team and Max availability due later.
Anthropic said it prioritises fixes by explaining severity, likely impact, reproduction steps and recommended remedies, while adding scheduled scans, dismissal records and CSV or Markdown exports.
The launch extends Anthropic's cyberdefence push after Project Glasswing, as the company tightens safeguards against misuse of AI vulnerability research and works with partners including CrowdStrike, Palo Alto Networks and PwC.
As AI finds bugs faster than humans can fix them, are we creating an unmanageable 'Patch Apocalypse'?
If AI now writes and secures our code, what is the next major vulnerability it will introduce?
Anthropic’s AI can hack better than humans. How long until attackers deploy their own AI weapons?