Company News: Oxford University Press Launches MCP Pilot; Partners Include Harvard Library, Wharton School, and University of Oxford

Here’s the Full Text of the Oxford University Press Announcement:

Oxford University Press (OUP) is working with three universities to securely connect their ChatGPT Edu accounts to academic content from OUP. Running from October to December 2026, the pilot will enable students and researchers to access OUP’s trusted content directly within their AI-assisted research journeys. Participating institutions include Harvard Library, the Wharton School of the University of Pennsylvania, and University of Oxford.

While researchers are increasingly integrating AI into their research journeys, a recent survey conducted by OUP shows their trust in AI outputs remains low. This pilot aims to ensure high-quality, trusted content remains discoverable and responds to changing habits brought on by AI, such as the rise in zero-click search journeys.

How the pilot works
The pilot uses a Model Context Protocol (MCP) server developed with OUP’s platform provider, Silverchair, which acts as a bridge between an AI assistant, such as ChatGPT Edu, and an organization’s content. In response to a user’s query, the MCP searches and retrieves the most relevant sources from the content, using them to formulate a response for the user.

Throughout the pilot, select researchers and students at the three institutions will be able to use ChatGPT Edu to query peer-reviewed content their institution has purchased or subscribed to from OUP. This means those AI outputs are grounded in trusted, cited scholarship with links to content hosted on the Oxford Academic platform.

What the pilot will inform
OUP and participating institutions will use both quantitative and qualitative data from the pilot to gain insight into researchers’ behaviours within AI research tools, including:

Whether researchers are motivated to use an MCP to connect to trusted sources How access to trusted sources influences the quality of the LLM’s outputs Whether the presence of citations from a trusted publisher affects the process of verifying resources.

Usage data will be monitored and analysed to assess the desirability and sustainability of publisher MCP connectors. This data, combined with qualitative feedback from faculty and students, will be used by each institution to inform the role MCPs will play in their future AI-powered research journeys.

Richard Ovenden, Bodley’s Librarian and the Helen Hamlyn Director of the University Libraries at the University of Oxford, said: “As people increasingly look to AI technologies to support their academic research, it is important that libraries and publishers work together to make their studies and research as reliable and trustworthy as possible. We are therefore constantly exploring new ways to improve access to the resources we make available to students and researchers. This collaboration will allow us to review how readers use AI technology in their studies and to help develop approaches to ensure that students and researchers can confidently rely on the information they discover.”

John Campbell, Product Strategy Director for OUP’s Academic division, said: “This pilot is part of a programme of initiatives and partnerships to understand how the dissemination of high-quality research adapts to new technologies and continues to drive impact. By integrating carefully-managed AI tools across our products, we’re putting researchers’ needs at the heart of our work, helping them harness the power of this evolving technology in a secure and reliable way.”

Tanya Laplante, Director of Product Platforms for OUP’s Academic division, said: “At a time of rapid technological change, it’s difficult to see through the hype to identify what will actually benefit researchers in their day-to-day work. We’ve undertaken this pilot to learn what real-world use of publisher MCPs looks like and what benefits it drives. This insight will help OUP and participating institutions determine where we go from here.”

Filed under: Companies (Publishers/Vendors), Data Files, Libraries, News, Publishing