Microsoft wants to limit AI agents’ permissions when they stray from the job

Microsoft announced Guardian Agents and new Playbooks as an extension of Project Perception, its agentic security system, designed to adapt as agent activity, risk, security context and organizational policy change. The Guardian Agents help security teams understand why risk changed, investigate relevant evidence, recommend or apply containment measures, and feed lessons back into policies, while people retain control over objectives and consequential decisions. Playbooks define desired security outcomes and coordinate protective actions.

Microsoft’s Guardian Agents in Project Perception adapt as agent activity, risk, security context and organizational policy change. They help security teams understand why risk changed, investigate relevant evidence, recommend or apply containment measures and use what they learn to strengthen policies. People retain control over objectives and consequential decisions. Microsoft announced the Guardian Agents and new Playbooks as an extension of Project Perception, its agentic security system. Playbooks define desired security outcomes and coordinate protective … More https://www.helpnetsecurity.com/2026/10/20/microsofts-ai-agents-threats-guardian-agents/ The post Microsoft wants to limit AI agents’ permissions when they stray from the job https://www.helpnetsecurity.com/2026/10/20/microsofts-ai-agents-threats-guardian-agents/ appeared first on Help Net Security https://www.helpnetsecurity.com .