- AI agents from startup Cathedral infiltrated Anthropic’s business systems in under 12 hours by exploiting third-party software vulnerabilities.
- Anthropic confirmed limited access to sales, HR, and IT systems but said core AI assets like Claude responses and model weights were untouched.
- The test prompted national security concerns, with officials warning that state actors could replicate the attack.
Researchers at Cybersecurity startup Cathedral used AI agents to find paths into Anthropic‘s internal business systems in less than 12 hours, according to a Washington Post report. The agents exploited weaknesses in third-party administrative and sales software commonly used by many companies.
Cathedral, founded by former staffers of the Department of Government Efficiency, tested Anthropic through a bug bounty program. Its AI agents accessed meeting transcripts, salary details, and Slack messages about Anthropic’s advanced models.
Anthropic confirmed that the researchers reached a limited set of sales, HR, and IT systems holding business contact details and basic user account records. However, customer conversations, Claude responses, source code, and model weights were not accessed.
Anthropic said Cathedral overstated the national security implications and did not fully disclose all its testing activity. Cathedral maintained that its data accesses were kept to the absolute minimum in line with security research best practices.
A senior Trump administration official said the report has circulated in government and raised concerns that state-sponsored actors could replicate the attack. This incident follows previous AI security issues, including Chinese Hackers using Claude Code to target organizations.
Anthropic disabled the affected software until vendors confirmed fixes.
✅ Follow BITNEWSBOT on Telegram, Facebook, LinkedIn, X.com, and Google News for instant updates.
