Story
arxiv_cs_ai ยท Nov 17, 2026 ยท paper
arxiv.orgNov 17, 2026
original source linked
In brief
Safety benchmarks for computer-use agents (CUAs) place the harm either in the intent of the task or on a single screen: a malicious instruction, an injected command, a phishing page. An agent that checks the task and...
Feed lens
agent