PromptArmor 发布企业 AI 智能体安全基准 FrontierBreach
PromptArmor has released FrontierBreach, a benchmark of how often AI agents at work can be misled by hidden instructions.
PromptArmor 发布 FrontierBreach,用于评测企业 AI 智能体工作流遭隐藏指令误导的频率。在21个 AI 模型的测试中,攻击成功率为34.9%;各前沿模型均被攻破,其中 Gemini Argon 为4.3%、Muse Wonder 为15.2%、Grok 4.5 为91.4%。隐藏在智能体技能中的攻击成功率为45.5%。
查看原帖今天,我们发布 FrontierBreach,这是衡量企业智能体工作流攻击成功率的一项新标准。 现有基准测试普遍低估了现代 AI 智能体面临的安全风险。 FrontierBreach 采用可能造成网络安全后果(勒索软件、数据外泄、网络钓鱼等)的复杂攻击链,模拟现实企业工作流中的攻击场景。
查看原文
Today we’re releasing FrontierBreach, a new standard for attack success rate on agentic workflows in the enterprise. Existing benchmarks systematically underrepresent the security risks facing modern AI agents. FrontierBreach reflects attack scenarios on realistic enterprise workflows, while using sophisticated attack chains that result in cyber consequences (ransomware, data exfiltration, phishing, etc).
正文
PromptArmor has released FrontierBreach, a benchmark of how often AI agents at work can be misled by hidden instructions.
Attackers can hide an invisible line in an email or document, and agents may try to follow it, sending data out or running the attacker's code.
> Across 21 AI models, these hidden attacks worked on 34.9% of PromptArmor's tests.
> PromptArmor breached every frontier model it tested, from Gemini Argon at 4.3% to Muse Wonder at 15.2%, with Grok 4.5 falling for 91.4% of the attacks.
> Attacks hidden inside agent skills also worked in 45.5% of the tests.
AI 热点