LesenForschungRadarAnlageframework
Anmelden / Registrieren
← Zurück zur Lektüre
NewsThema · 思科AI智能体测试Kurzfassung2 Min. LesezeitDisconfirmAI Redaktion

Cisco AI Security Team: Generic Test Banks Cannot Find the Risk of Agents Being Exploited Through Normal Function

The Cisco AI security product team points out that AI agents hold credentials, call tools, and write to systems of record, making them no longer static features inside applications. A generic test bank can only find errors common to all agents; it cannot find the unique failure modes that arise because each agent has different permissions, tools, and purposes—for example, a support agent talked into refunding the wrong person is not a jailbreak but the agent doing its job.

2026-10-21 04:59 GMT+8

Agents are no longer just features inside applications. They hold credentials, call tools, and write to systems of record. This means their security testing cannot follow the logic of traditional software scanning: run a test suite, read the results, fix what it found.

The Cisco AI security product team, in a blog post dated October 20, 2026, drew a specific distinction: a support agent talked into refunding the wrong person is not a jailbreak. It is the agent doing its job, except the beneficiary is now an attacker. A test bank written for every agent finds only what is wrong with every agent; it will not find the error unique to this particular agent.

This distinction explains why testing must start from context. A healthcare scheduling agent and a financial advisory agent fail in different ways, because they serve different purposes, reach different tools, and carry different permissions. The adaptive red-teaming process Cisco describes works as follows: first study the agent's purpose, the tools it can reach, and the guardrails already in place; then derive the outcomes that agent must never produce, and pursue them across many turns, pivoting when one approach is blocked. The question stops being whether the agent fails a known attack and becomes what an attacker who reads your documentation would try.

Anmelden / Registrieren
LesenForschungRadarAnlageframework

Cisco also noted that standardized testing and adaptive testing each have a role: the former provides breadth against public frameworks (OWASP Top 10, MITRE ATLAS, NIST), yielding wide coverage, comparable results, and something an auditor can repeat; the latter provides depth on the agents that matter most. No one can hand-write a test plan for every agent in an enterprise, but if the goals are derived from each agent's own context, no one has to.

It should be noted that the above is Cisco's own description of its product methodology, with no independent verification data, comparative experiments, or third-party evaluation results attached. The Cisco–Workday partnership—Workday Agent Passport is currently in Early Access, with Cisco AI Defense serving as its launch assessor—is the first public practice case of this method, but its actual security effectiveness has not yet been independently verified.

Quellen

  • blogs.cisco.com↗
Thema

思科AI智能体测试

思科(Cisco)指出,通用测试库无法发现特定AI智能体因权限和工具差异而产生的独特风险,如客服智能体被诱导执行错误退款。为此,思科推出“自适应红队测试”,并作为Workday Agent Passport的评估伙伴,为智能体提供基于OWA…

5 Schritte1 BeiträgeAktualisiert 2026-10-22 01:56

Wie es dazu kam

  1. 2026-05-04思科发布AI Defense Explorer Edition实验室指南,强调红队测试应前置到开发阶段,支持自定义目标以发现隐藏指令泄露等风险。blogs.cisco.com ↗
  2. 2026-06-01思科宣布AI Defense重大更新,引入“自适应红队测试”和Policy Studio护栏,针对每个智能体的独特上下文进行定制化安全保护。blogs.cisco.com ↗
  3. 2026-06-16思科在Explorer Edition中推出Agent Validation功能,专门测试智能体框架特有的攻击面,如工具路由滥用和持久状态修改。blogs.cisco.com ↗

Was noch ungeklärt ist

  • 其他主要智能体平台是否会采纳类似的第三方独立验证+安全印章模式?
  • 自适应红队测试在实际大规模部署中的误报率和漏报率具体数据如何?
  • Workday Agent Passport从早期访问到全面正式发布的过渡时间表及客户采用情况。
Vollständiges Thema lesen →
Recherche fortsetzen →