Tech
AI Research Raises Concerns About Anthropic's Claude
Security researchers have uncovered vulnerabilities in Anthropic’s AI, Claude. The findings suggest that the company’s carefully crafted persona may be a potential risk.
Coverage timeline — 1 article
The Verge
Anthropic has spent years building itself up as the safe AI company. But new security research shared with The Verge suggests Claude's carefully crafted helpful personality may itself be a vulnerability. Researchers at AI red-teaming company Mindgard say they got Claude to offer up erotica, maliciou
2026-05-05 13:13 UTC