OpenAI Adds Support for Anthropic’s Model Context Protocol (MCP)
OpenAI CEO Sam Altman Announces Support for MCP
OpenAI CEO Sam Altman announced on Wednesday that the company’s developer team has added support for the Model Context Protocol (MCP), an open-source standard for connecting artificial intelligence applications to data repositories. This allows developers to use MCP to ensure their chatbots only provide responses based on relevant information.
According to the company’s developer team, the protocol can now be used in the Agents SDK, its toolkit for building AI-powered agents. The team is also working on adding support for the ChatGPT desktop app and OpenAI API, and will announce more in the coming months.
SEE: Have Some Spare Cash? You’ll Need it for OpenAI’s New API
How Does MCP Help AI Systems Access Data?
Scaling AI models can be a clunky process, as they constantly need access to new third-party data sources, such as cloud servers, business platforms, and legacy systems. Each source will speak its own technological language, requiring developers to build custom integrations for each one.
MCP is effectively a universal language that can be used to develop two-way connections between a wide range of AI models and third-party data sources. “The result is a simpler, more reliable way to give AI systems access to the data they need,” Anthropic said.
More Must-Read AI Coverage
Developers creating AI agents with Agents SDK can either expose their data through so-called “MCP servers” or build “MCP clients,” i.e., apps that easily connect to the servers. This removes the need for countless custom-built integrations and allows AI to access the information it needs across different tools without losing context.
Anthropic Chief Product Officer Mike Krieger said he was “excited to see the MCP love spread to OpenAI” in an X post. “LLMs are most useful when connecting to the data you already have and software you already use.”
MCP was made open-source in November and since then a number of AI platforms have announced support, including Microsoft Copilot, Apollo, Replit, Codeium, and Sourcegraph.
Which AI Platforms Support MCP?
It was just over a week ago that Microsoft announced MCP support in Copilot Studio, allowing users to connect their agents to MCP servers directly within the platform. But just after the OpenAI announcement, Redmond also released a Playwright MCP server on GitHub. MCP agents will easily connect to the server to gain automated browser capabilities such as clicking, typing, and browsing the web.
Naturally, the more AI companies that adopt MCP, the more valuable it becomes for the AI community. With OpenAI — Anthropic’s biggest rival — now on board, Anthropic may be on its way to establishing the global standard for agent interconnectivity.
Conclusion
In conclusion, OpenAI’s addition of support for the Model Context Protocol (MCP) marks a significant step forward in the development of AI-powered agents. By providing a universal language for connecting AI models to data repositories, MCP has the potential to simplify the process of scaling AI models and improve the overall efficiency of AI development.
FAQs
- What is the Model Context Protocol (MCP)? MCP is an open-source standard for connecting artificial intelligence applications to data repositories.
- What are the benefits of using MCP? MCP provides a simpler, more reliable way to give AI systems access to the data they need, removing the need for countless custom-built integrations.
- Which AI platforms support MCP? A number of AI platforms have announced support for MCP, including Microsoft Copilot, Apollo, Replit, Codeium, and Sourcegraph, as well as OpenAI.
- What is the potential impact of OpenAI’s addition of support for MCP? OpenAI’s addition of support for MCP has the potential to simplify the process of scaling AI models and improve the overall efficiency of AI development, as well as potentially establishing the global standard for agent interconnectivity.
