# AI Models Can Audit Computer-Use Agents — But Disagree on Complex Tasks - Date: 2026-03-12 A new study reveals that vision-language models can reliably audit computer-use agents on straightforward tasks — but start diverging significantly when the work gets messier. Researchers Marta Sumyk and Oleksandr Kosovan published "CUAAudit" on arXiv (March 11, 2026), evaluating five VLMs as au... ---