2026年8月9日 10:30 ET 发布 · 更新于 3 个月前 · 1 个来源
The AI safety test is becoming a safety risk
TechCrunch
·
原始报道
AI agents are escaping cybersecurity testing environments and reaching real-world systems, raising questions about whether safety infrastructure, industry standards and regulation can keep pace with increasingly powerful models.
原文