Broadcom Launches AgentMinder for AI Governance
Broadcom has introduced AgentMinder, a new enterprise solution designed to govern autonomous AI agents and control their actions at runtime. The company announced the platform on August 31, 2026, at VMware Explore 2026 in Las Vegas.
As AI agents move beyond content generation, they increasingly access enterprise data, invoke tools and execute business processes. Therefore, Broadcom says companies need controls that operate continuously rather than relying only on model guardrails or static permissions.
AgentMinder Adds Identity and Runtime Controls
AgentMinder treats autonomous AI agents as enterprise-grade identities. Moreover, it links each agent’s authority to a defined mission, permitted intents, approved tools, and authorized resources.
The platform also checks an agent’s identity and intent before allowing an action to reach an enterprise resource. In addition, its cloud-native AI gateway evaluates each tool call against policies and current context.
Broadcom says the system can enforce least-privilege policies during runtime. Consequently, organizations can restrict agents from taking actions beyond their approved authority.
Observability and Enterprise Deployment
Meanwhile, AgentMinder adds an observability layer built on OpenTelemetry. As a result, security and platform teams can track agent sessions and actions while maintaining audit trails and visibility into anomalies.
The platform can run alongside existing large language models across on-premises systems, private clouds and public clouds. Furthermore, it supports integration with existing authorization systems through the AuthZEN standard.
Broadcom says AgentMinder already handles nearly 36 million customer-related API calls and seven million workforce-related API calls each day. It also supports more than 20 million customer identities and 72,000 workforce identities.
Broadcom Targets Wider Agentic AI Adoption
The company has made AgentMinder generally available. In addition, Broadcom plans to position it within a wider security architecture for agentic AI.
That architecture includes AgentMinder, VMware vDefend, and VMware Avi Load Balancer. Together, these technologies target identity governance, runtime enforcement, workload protection and threat detection.
Therefore, Broadcom is positioning AgentMinder as a control layer for enterprises moving autonomous AI agents into sensitive workflows, including finance, HR and IT.
