CVE 报告量飙升至 1,500 —— AI 漏洞狩猎才刚刚开始
2026 年 6 月，随着 claude mythos 等 ai 模型自动化漏洞发现，cve 报告量达到创纪录的 1,500 份。这大量的漏洞表明，ai 扫描正成为安全基线，标志着根本性的转变。
security vulnerability reports hit an all-time high of 1,500 in june 2026, more than 3.5 times the previous monthly record. epoch ai attributes the spike to widespread deployment of ai models like anthropic's claude mythos preview and its glasswing initiative, which alone has identified over 10,000 high-severity flaws. competitors like openai are likely contributing to this wave of automated discovery, fundamentally changing the speed at which the internet is secured. the explosion in reports suggests that automated scanning is becoming essential for maintaining software safety.