Microsoft adds Guardian Agents and Dynamic Permissions to police workplace AI agents
Microsoft extended its Project Perception security work to cover AI agents inside companies, adding Guardian Agents that investigate and contain risky agents, task-scoped Dynamic Permissions, and discovery for 60 kinds of local agents such as Claude Code and Codex.
What happened
- On October 19, 2026, Microsoft said it is extending Project Perception to treat AI agents as a new kind of insider that needs its own security controls.
- Discovery now covers 35 more types of local agents, 60 in total, including Claude Code, OpenAI Codex and GitHub Copilot, which can be brought into one registry.
- New Dynamic Permissions give an agent its own task-scoped, time-bound access instead of inheriting everything its user can do, with human approval for higher-impact actions.
- Dynamic Agent Risk and Dynamic Conditional Access adjust what an agent may do as its risk changes, from blocking writes to stopping it entirely.
- A team of Guardian Agents, guided by playbooks and rules people set, investigates why an agent's risk changed and recommends or applies containment.
Why it matters: If your team lets AI agents touch company email, files or finance systems, this is a sign that agent permissions and monitoring are becoming a standard part of IT security. Small businesses on Microsoft 365 should expect these controls to show up in Entra and Defender.
Sources
- Microsoft Security Blog: Agents are your newest insiders microsoft.com
- TechCrunch techcrunch.com
Checked October 20, 2026 against the sources above. Claims about performance or benchmarks are the company's own unless stated otherwise.