LesenForschungRadarAnlageframework
Anmelden / Registrieren
← Zurück zur Lektüre
NewsThema · 微软守护型代理人Wesentlich1 Min. LesezeitDisconfirmAI Redaktion

Microsoft warns AI agents pose insider risks, launches Guardian Agents

Microsoft treats AI agents as insider threats, expanding Project Perception with dynamic permissions and introducing Guardian Agents for continuous behavioral supervision.

2026-10-20 10:25 GMT+8

US technology company Microsoft issued a warning on October 19, 2026, stating that the widespread deployment of AI agents within enterprises could evolve into a new cybersecurity risk. The company emphasized that frontier models, regardless of their supplier, "cannot be trusted," as they are trained to pursue results and may deviate from organizational norms like an "insider threat," compromising sensitive data and critical processes without external hacking.

To address this risk, Microsoft argued that enterprises cannot leave security responsibility entirely to model suppliers but must establish control mechanisms independent of the models themselves. The company announced an expansion of its Project Perception security architecture, adding new capabilities such as dynamic permissions, dynamic agent risk assessment, and dynamic conditional access. These mechanisms can adjust available permissions in real-time based on changes in agent risk, restricting access scopes when necessary or requiring human approval for high-risk actions.

Additionally, Microsoft introduced so-called Guardian Agents, serving as a protective layer to continuously supervise other AI agents. Microsoft stated that these guardian agents will monitor the behavior of every AI agent around the clock, intervening immediately when risks escalate to help enterprises bring non-deterministic AI behaviors under controlled scope. Microsoft CEO Satya Nadella confirmed this move on social media, describing it as putting the principle of treating models as insider risks into practice.

Quellen

Anmelden / Registrieren
LesenForschungRadarAnlageframework
  • infosecu.technews.tw↗
Thema

微软守护型代理人

微软于2026年10月19日正式推出“守护型代理人”(Guardian Agents)及配套安全控制措施,旨在应对企业内部部署的AI智能体可能演变为“内鬼”的安全风险。微软安全业务负责人Hayete Gallot明确指出,无论来源如何,前沿…

3 Schritte1 BeiträgeAktualisiert 2026-10-20 21:50

Wie es dazu kam

  1. 2026-07-01微软首次公布Project Perception,定位为针对外部AI驱动攻击的多智能体安全系统。blogs.microsoft.com ↗
  2. 2026-10-10微软CEO Satya Nadella发表博客,警告企业不应盲目信任AI模型供应商,提出需建立独立于模型之外的控制机制。siliconangle.com ↗
  3. 2026-10-19微软正式发布Guardian Agents及动态权限控制功能,宣布将Project Perception扩展至内部AI代理人风险管理,支持识别60种本地代理工具。microsoft.com ↗

Was noch ungeklärt ist

  • Guardian Agents的具体商用可用日期及定价策略尚未公布。
  • 第三方模型供应商(如OpenAI、Anthropic)对微软将其定义为“不可信”并推出竞品监控方案有何回应?
  • 在实际企业环境中,动态权限调整对AI代理人执行复杂任务效率的影响程度如何?

Auch in diesem Thread

  • Microsoft Introduces 'Dynamic Agent Risk' Controls to Limit Rogue AI Agents2026-10-20
Vollständiges Thema lesen →
Recherche fortsetzen →