AI Agents Don’t Get Hacked. They Get Authorized. The New Vulnerability Class With No CVE
中文摘要
AI智能体面临新型安全风险：其漏洞不在于被黑，而在于权限被滥用，形成了一种无需CVE的新型授权漏洞类别。
English Summary
AI agents face a new vulnerability class: they aren't hacked, but rather exploited via improper authorization, bypassing traditional CVE classifications.
Original Excerpt
In May 2025, a security team in Zurich did something almost boring. They wrote a GitHub issue. That’s just a message on GitHub, the kind… Continue reading on Medium »