default author photo

Karolina Korgul

University of Oxford - Oxford Internet Institute

Graduate Student

SCHOLARLY PAPERS

1

DOWNLOADS

17

TOTAL CITATIONS

0

Scholarly Papers (1)

1.

Agent Benchmarks Fail Public Sector Requirements

Number of pages: 24 Posted: 12 Feb 2026
University of Oxford - Oxford Internet Institute, Hertie School - Centre for Digital Governance, University of Oxford - Oxford Internet Institute, Weizenbaum Institute Berlin and University of Oxford - Oxford Internet Institute
Downloads 17 (1,518,554)

Abstract:

Loading...

Large Language Model, AI Agents, Public Sector, Evaluation, Responsible AI, Agentic AI