小瑶速读：这一周，AI 的战场从模型“卷参数”切到了算力与交付：老黄把首批 Vera CPU 锁给 OpenAI/Anthropic/SpaceX；三星×OpenAI 定制芯传停滞、三星转投 Anthropic；微软、GitHub Copilot、Codex 与机器人线同步生变：◈Codex Windows端正式支持Computer Use功能，远程接管桌面任务OpenAI于5月29日更新Code ...

ToolCUA 的核心价值在于指出了 CUA 训练中的一个关键转折：当 Agent 从 GUI-only 进入 hybrid action space 后，能力瓶颈从“能否看懂界面”进一步变成“能否编排多种动作路径”。 这个问题看起来答案应该是肯定的 ...

SaaS-Bench用23个开源SaaS系统、106个任务测试Agent，结果全军覆没，暴露其在真实环境中的四种致命缺陷，距真正替人干活尚远。 想象一个真实的工作日：项目经理要更新项目状态，财务人员要整理客户账单，医疗管理员要核对预约和保险信息。 这些并不是高级 ...

OpenAI announced today that Codex app users on Windows 11 now have computer use capabilities and ChatGPT mobile app integration.

Forbes contributors publish independent expert analyses and insights. Tor Constantino is an ex-reporter, turned AI consultant & tech writer. Anthropic launched two updated versions of its Claude AI ...

Codex Desktop expands from coding into full productivity workflows. Automation can generate images, charts, and workflow outputs. The tool is still aimed at developers despite the broader productivity ...

Claude Code上线Computer Use，直接捅破开发效率天花板。 在官方演示中，只甩一个指令过去，AI就自己启动正在开发的应用，自己复现bug，自己修复，自己测试修复效果。 相当于直接给每个开发者配了个全能测试工程师。 这已经是Anhtropic在60天里的第76个更新。 与上周更新的桌面端Computer Use不同，CLI端更适合和现有开发工作流集成。 能完美融入开发者现有的命令行 ...

Imagine an AI model that can work with a computer all on its own. Well, imagine no longer because such an AI has arrived. On Tuesday, Anthropic announced that the latest generation of its Claude AI ...

$200 per month to "operate my computer"? Unless I have a remote only job and it can do my job for me I can't see why anyone would do that.

AI technology is advancing from chat-based assistants to autonomous agents capable of operating computers, executing multi-step tasks, and working across applications. Recent launches like Anthropic's ...