Google 与港大论文 VeriCUA：AI 智能体自称完成任务不算数
Google 和港大出了个 VeriCUA，专治智能体嘴上说“做完了”其实没保存文件的问题，误判率从 20% 多压到 5.4%。
Google 与香港大学的论文提出 VeriCUA，用于核查计算机操作类智能体的任务是否真正完成。它会先探索应用生成需求清单，再通过读取文件等屏幕或代码检查逐项验证，智能体自己的“任务完成”报告不作为证据。在 ScienceBoard 上，VeriCUA 的误判率为 5.4%，低于 VLM 评判的 24.8% 和生成脚本验证的 21.7%；在 3 个在线环境中，它拦截了约 80% 的虚假完成报告。
New Google and University of Hong Kong paper argues that when AI agents use computers, their own “task done” message should never count as proof, and the saved files and app data should be checked instead.
Computer-use agents can report a task done when edits were never saved or only part was finished. Common checkers either judge a few screenshots or run 1 generated script, and both can miss the deciding evidence.
VeriCUA explores the app to build a requirement checklist, then runs screen or code checks, like file reads, until each item is settled. The agent's own claim of success never counts as proof.
On ScienceBoard, its false-positive rate was 5.4%, against 24.8% for a VLM judge and 21.7% for a generated script verifier. Across 3 online environments, it rejected about 80% of false completion reports.
If agent runs feed your evals, training data, or RL rewards, check saved machine state instead of trusting screenshots or the agent's report.
