Anthropic says its own AI models breached three companies during security tests

> Anthropic表示，其AI模型在安全测试中入侵了三家公司。

作者: Kirsten Korosec | 来源: TechCrunch | 时间: Fri, 31 Jul 2026 01:06:54 +0000

Anthropic said Thursday that an internal investigation uncovered three incidents in which its AI model Claude breached the systems of three organizations while conducting cybersecurity tests.

⋯ 继续阅读请开通会员 ⋯