Kiteworks has announced a strategic partnership with Reco aimed at giving enterprises greater visibility and control over AI agents, SaaS applications and identities as organizations expand their use of agentic AI.

The partnership is designed to address what the companies describe as a growing governance gap as AI agents are adopted more quickly than security teams can review them. By combining Reco's agent and identity discovery capabilities with the Kiteworks control plane, organizations are intended to gain visibility into where agent risk exists while governing what AI agents can access, edit, share and exchange through a single audited policy.

Reco continuously discovers and maps SaaS applications, human and non-human identities and AI agents operating across an enterprise, identifying application sprawl, misconfigurations and excessive permissions. That intelligence is designed to feed into the Kiteworks platform, which governs how sensitive data is accessed, edited, shared and exchanged across email, file sharing, APIs and AI agent workflows.

According to Kiteworks, its Data Policy Engine can respond to risk signals in real time by applying view-only access, SafeEDIT possessionless editing, encryption or blocking access entirely. The company's Secure MCP Server extends the same governance model to AI agents built on the Model Context Protocol, applying the same policies, encryption and audit trails used for human users.

The announcement also cites Kiteworks research indicating that enterprise AI governance remains limited. According to the company, fewer than 31% of organizations have deployed AI containment controls. The research also found that 65% of organizations identified employees using unapproved AI tools with organizational data during the past year while fewer than half implemented technical controls afterward to prevent repeat incidents.

For organizations in regulated industries, the companies said the partnership is intended to help address compliance requirements by providing audit-ready evidence that spans identity discovery through data activity. Kiteworks said its platform supports compliance with frameworks including HIPAA, CMMC 2.0, GDPR and PCI DSS.

"Security teams don't have a visibility problem anymore. Tools like Reco have solved that," said Sean Kelley, global director of strategic alliances at Kiteworks. "What they've had is an enforcement gap – specifically, knowing an agent exists and what it can reach is not the same as governing what it does with sensitive data. This partnership connects those two halves. Reco tells you where the risk is, and the Kiteworks control plane makes sure that risk is governed the moment it touches sensitive data, whether the identity behind it is a person or an agent."

Zoe Hillenmeyer, chief operating officer at Reco, said enterprises are adopting AI agents across third-party ecosystems faster than they can govern them. She said the partnership gives customers visibility into every agent and identity interacting with their environments while providing a way to enforce data governance over what those agents can access and do.

Kiteworks is showcasing its Compliant AI capabilities at AI4 2026, where Craig Pfister, global vice president of solutions architecture, is presenting a session titled "Controlling Data Access and Use by AI Agents for Compliant AI." Reco is also exhibiting at Black Hat USA.