YOUR GO-TO INTELLIGENCE RESOURCE FOR THE SCHOLARLY PUBLISHING INDUSTRY

STM and Ithaka S+R have launched a new research project examining how scholarly content should be accessed and used by AI-powered research assistants and agentic systems.

The initiative, titled “The Endpoint Question,” will investigate how AI agents discover, access, use, and attribute scholarly content. The study aims to gather evidence that can help shape future standards and implementation approaches for trusted interactions between AI systems and scholarly resources.

The project forms part of TRACE, a collaborative initiative involving STM, NISO, COUNTER, and other stakeholders focused on trusted retrieval and attribution in AI-mediated scholarly communication.

The research builds on STM’s recent consultation on the responsible use of research content in generative AI, which highlighted broad support across the scholarly communications community for principles such as verifiable attribution and citation, transparency, inclusion of corrections and retractions, and clear distinctions between peer-reviewed and non-peer-reviewed content. The new study seeks to explore how these principles can be implemented in practice and what a trusted scholarly endpoint for AI systems should look like.

Led by Ithaka S+R, the three-month study will examine how AI agents identify scholarly content, authenticate through user entitlements, receive provenance information, and create reliable records of content usage. The research will bring together publishers, repositories, infrastructure providers, and AI developers to assess emerging approaches, identify challenges, and define requirements for future implementation.

Ithaka S+R will conduct the study independently and plans to publish its findings openly. Areas of investigation will include:

• Defining a trusted scholarly endpoint for AI agents

• Understanding and identifying agent-mediated activity

• Ensuring provenance and attribution information is maintained throughout AI workflows

• Measuring and evaluating scholarly content consumption by AI systems

• Assessing approaches most likely to be adopted by AI developers and platform providers

Organizations interested in participating in the study can contact representatives from STM Solutions or Ithaka S+R for further information.