Agentic AI in University Administration

A shared task at the CIKM 2026 AnalytiCup, Rome, 8–11 November 2026.

Synopsis / Call

Agentic AI systems combine large language models, information retrieval, planning, and tool use to act on behalf of users in complex digital environments. Their evaluation, however, remains difficult for lack of reproducible, bounded laboratories. A university administration is a closed-world, socio-technical environment whose tasks follow standardized, well-documented processes — making it particularly suited for a Cranfield-style evaluation of AI agents. The controlled, reproducible evaluation of agentic AI in such enterprise-like settings is an underexplored yet highly relevant problem.

This shared task offers the first reusable laboratory for the controlled evaluation of agentic AI in a university environment, namely the University of Kassel. Participants build agents that solve standardized administrative tasks (e.g., travel permits and reimbursements) by retrieving the relevant regulations, planning the necessary steps, and using a controlled set of tools, while complying with the institution's codes of conduct. The task runs as part of the CIKM 2026 AnalytiCup and is organized together with the OpenWebSearch.eu project.

For the full task definition, data, evaluation, and participation process, see the Task page.

Shared Task

Important Dates

All dates are tentative and will be finalized upon acceptance.

  • July 2026: Launch; release of collection, tools, and example topics; registration opens.
  • August 2026: Team formation deadline; training topics released.
  • Mid-October 2026: Submission deadline for system runs.
  • Late October 2026: Evaluation period.
  • Early November 2026: Notification of results.
  • 8 November 2026: Presentation at the CIKM 2026 AnalytiCup (main conference 9–11 November).

The timezone of all deadlines is Anywhere on Earth (AoE).

Sponsors

To be announced.