OpenAI is set to integrate support for Anthropic’s open-source Model Context Protocol (MCP) across its suite of products. The announcement was made by CEO Sam Altman on Wednesday, revealing that the Agents software development kit (SDK) now includes MCP support, with additional products anticipated to follow suit shortly. MCP aims to standardize the process by which AI systems link to external data sources, such as third-party repositories and data hubs, ensuring uniform behavior among chatbots.
OpenAI to Integrate MCP Across Its Product Line
In a tweet on X (formerly Twitter), Altman expressed the company’s commitment to adopting the protocol, highlighting its popularity among both users and developers. Currently, developers can access MCP support within the Agents SDK, and plans are in place for two additional products to add this functionality in the near future.
The forthcoming products to incorporate MCP are the ChatGPT desktop application and the Responses application programming interface (API). Additionally, OpenAI is reportedly developing a feature that will enable Team subscribers to connect the chatbot with Google Drive and Slack. It is expected that this feature will also embrace MCP once launched. OpenAI noted that further details will be shared in the upcoming months.
Anthropic made the MCP available to the AI community last year, addressing a significant hurdle for AI systems. Many chatbots driven by large language models (LLMs) primarily depend on internal databases to respond to user inquiries.
However, situations may arise where an AI needs to answer questions based on external knowledge, necessitating connections to third-party cloud servers and various data repositories. The current landscape of AI technology is still evolving, leading to disparate systems that employ unique methods for interfacing with data hubs.
This discrepancy results in AI chatbots exhibiting varied behaviors when accessing these datasets. Challenges can include latency issues, differences in accessible file formats, and variations in how the model processes information. MCP addresses these concerns by providing a unified protocol for linking to external data hubs.
In response to the announcement, Mike Krieger, Chief Product Officer at Anthropic, shared his enthusiasm in a post, stating, “Excited to see the MCP love spread to OpenAI! MCP has [become] a thriving open standard with thousands of integrations and growing. LLMs are most useful when connecting to the data you already have and software you already use.”
