AI 能看界面也能写代码，但合起来还差得远
现在的 AI 助手分两派：一派会点鼠标、看屏幕，另一派会写代码、敲命令行。真实工作两样都得会，还得随时切换——先看看界面长啥样，再写段代码，再跑起来亲眼确认结果对不对。这篇论文造了个训练场：给 AI 一个正在运行的软件当参照，让它自己摸索行为、自己写实现，没人告诉它步骤。五个平台(Windows、macOS、Ubuntu、安卓、网页)都能跑，还配了 250 道考试题。最强模型 GPT-6 Astra 总分 58.1%，但只有 2.8% 的任务能全对——也就是说，它大部分时候是「差不多会」，不是「真会」。AI 复制静态界面比复制交互和计算结果更靠谱，生成的软件也比原版更小、更笨重。这不是你明天能用的东西，但它划了一条线：能看、能写、能验证，三件事合在一起，才是真干活。
📄 原文摘要(英文)
Computer-use agents (CUAs) have advanced along two separate lines: graphical interaction and software development through code and the command line. Real digital work requires both, interleaved rather than stacked end to end. We study hybrid CUAs that autonomously decide when to explore an interface, implement software, and run and visually verify their artifacts. We introduce RecreationWorld, a five-platform framework built around recreation: given a running reference, an agent must discover its behavior and build a faithful implementation with no prescribed workflow. RecreationWorld provides reproducible environments on Ubuntu, macOS, Windows, Android, and Web, plus a unified harness with native GUI control and coding tools. The running reference serves as an oracle for hidden behavioral tests, providing execution-grounded rewards. We scale trajectory generation with high-quality open-source applications. Models trained on these trajectories improve across five out-of-distribution coding and hybrid computer-use benchmarks and more frequently verify their rendered outputs, providing evidence of transfer beyond recreation. For held-out evaluation, we introduce RecreationBench, comprising 250 diverse tasks across domains and platforms. Reference-grounded programmatic and visual assertions cover action-conditioned outcomes at multiple interaction depths; each is validated on the reference and by human reviewers before the suite is frozen for automatic scoring. GPT-6 Astra leads at 58.1% overall, but passes all programmatic tests on just 2.8% of tasks. Agents reproduce static interface structure more reliably than interactions and computed outputs, while generated applications remain smaller and more monolithic than their references. We release the benchmark, environments, and test suites.