OpenAI has announced its adoption of Anthropic’s standard for integrating AI assistants with the systems that store data.

In a recent post on X, OpenAI CEO Sam Altman revealed that the company will be incorporating support for Anthropic’s Model Context Protocol (MCP) across its range of products, including the ChatGPT desktop application. MCP is an open-source standard designed to enhance the relevance and accuracy of AI model responses to specific queries.

According to Altman, “The community has shown enthusiasm for MCP, and we are eager to integrate it across our products.” He further stated, “[Support for MCP] is now available in the Agents SDK, and we will be adding it to the ChatGPT desktop app and Responses API soon.”

The Model Context Protocol enables AI models to access data from various sources, such as business tools, software, content repositories, and app development environments, to complete tasks. This protocol facilitates the creation of two-way connections between data sources and AI-powered applications, including chatbots.

Developers can utilize MCP by setting up “MCP servers” to expose data and building “MCP clients,” such as apps and workflows, that connect to these servers as needed. Since Anthropic open-sourced MCP, several companies, including Block, Apollo, Replit, Codeium, and Sourcegraph, have integrated MCP support into their platforms.

Anthropic’s chief product officer, Mike Krieger, expressed his excitement about OpenAI’s adoption of MCP, stating, “Thrilled to see OpenAI join the MCP community – welcome!” He emphasized that MCP has become a thriving open standard, with thousands of integrations and growing, and that large language models are most useful when connected to existing data and software.

OpenAI plans to share more details about its MCP implementation in the coming months, as announced in a recent X post.