
The latest challenge? Model context protocol, or MCP, servers, which allow AI agents to connect to data and tools via a natural-language interface.
“With VCF 9.0, we already enabled an AI agent builder service which helps enterprises to easily build AI agents,” said Prashanth Shenoy, vice president of product marketing for the VCF division at Broadcom, at last week’s press conference. “In the future, we will support MCP, the model context protocol, for providing enterprises with a secure method to integrate their AI assistant with external content, repo and tools like GitHub, SQL Server, Oracle ServiceNow, Postgres and many more, without them having to build and maintain custom connectors.”
The tech preview of some of this new functionality was officially announced today.
For example, customers of VMware’s Avi Load Balancer will be able to secure their MCP traffic with WAF, while also providing MCP session persistence and authorization to help reduce the new attack surface.
Both session persistence and authorization are challenges for enterprises trying to deploy MCP servers on their own, since the base protocol is still new and lacking in control features.
“MCP is the protocol of choice for agentic AI,” said Umesh Mahajan, general manager for application networking and security at Broadcom. Support for the protocol allows enterprises to participate in the broader MCP ecosystem, he said. “The world is moving to MCP,” he said.