Identity security organization Silverfort has announced it will integrate its Identity Security control for AI agents into Microsoft Copilot Studio.
Runtime security addresses Copilot actions before execution
The integration will enable Silverfort to deliver inline identity security at runtime, enforcing intelligent access control policies the moment a Copilot agent attempts to act, blocking unauthorized access before execution.
“The more access an AI agent has to corporate resources, the more powerful it becomes. Without deep identity context, there’s no way to make an informed, real-time decision about whether an agent’s action is legitimate or overreach. That’s why agentic security is an identity problem at its core,” said Ron Rasin, chief strategy officer at Silverfort.
“Silverfort’s integration with Microsoft Copilot Studio is a recognition that runtime identity enforcement isn’t optional, it’s the foundation for deploying AI with confidence.”
AI agents created in Copilot Studio authenticate, retrieve enterprise data, trigger workflows, and interact with enterprise systems across cloud and on-prem environments.
Microsoft reports that over 80 percent of Fortune 500 organizations are deploying active agents built with low-code/no-code tools. Additionally, 29 percent of employees already use unsanctioned AI agents for work.
Utilizing the integration for runtime enforcement
Silverfort will integrate into the Microsoft Copilot Studio ecosystem for runtime enforcement. Silverfort will evaluate the request in real-time and return a decision before the action executes each time a Copilot agent requests access to a tool or function.
This capability will prevent unauthorized access, privilege escalation, and unintended actions before they occur, enabling organizations to adopt AI agents with confidence.
Silverfort’s runtime enforcement features:
- Limits unauthorized privilege elevation: AI agents cannot exceed authorized privileges.
- Stops risky AI agent overreach: Blocking anomalous access attempts before execution.
- Dynamically adapts access policies: Granular access policies dynamically adapt based on real-time context and risk.
- Creates audit trails: All activity is auditable and tied to enterprise identity governance frameworks and the human using the agent.
“Silverfort’s integration with Microsoft Copilot Studio extends security into the moment of access itself,” said Ankur Arora, Principal Group PM Manager, Microsoft. “This isn’t visibility after the fact. It’s inline, runtime enforcement that evaluates and controls every agent’s access requests before they execute.”
Identity security remains a focus for entire Silverfort ecosystem
Further, Silverfort will extend Identity Security beyond Copilot Studio to deliver unified visibility and control over Copilot Studio AI agents; human identities; non-human identities, including service and machine accounts; and external and third-party AI agents operating outside the Microsoft ecosystem.
As a long-standing Microsoft collaborator and former Microsoft Partner of the Year, Silverfort will continue to work closely with Microsoft to develop AI-focused security features and expand its role in protecting hybrid and cloud environments.
Recently, Microsoft has begun to emphasize operationalizing AI through copilot, agents, Agent 365, and Microsoft 365 E7 as they move into production. Learn more about the company steering partners toward AI as part of Microsoft’s next phase of AI, which it’s calling the Frontier Transformation.