Google宣布預覽其最新一代AI模型「Gemini 2.5 Computer Use」，這款模型不只能理解文字或影像，還能像人一樣實際「操作」網頁介面。透過點擊、捲動、輸入與拖曳等動作，AI能在沒有API串連運作情況 ...

Google 於 10 月 7 日發布 Gemini 2.5 Computer Use 模型，這款基於 Gemini 2.5 Pro 視覺理解和推理能力開發的專門模型，讓 AI 代理能夠像人類一樣操控用戶介面。 該模型在多個網頁和流動裝置控制基準測試 ...

Google 推出 Gemini 2.5 Computer Use 模型，使 AI 代理能以視覺理解操作圖形介面 (GUI)，執行點擊、輸入等網頁任務。開發者可透過 Gemini API 取用，模型已內建安全控制機制，確保高風險操作的安全性。#Google,人工智慧,UI,AI Agent,Gemini API,Gemini 2.5 Computer Use ...

Google現開放用戶透過Gemini API試用Gemini 2.5 Computer Use預覽版，以人工智慧代理在網頁上執行點擊、輸入與拖曳等，並內建安全審查與使用者確認機制，阻擋惡意使用與意外行為等操作 熱門新聞 ...

Anthropic has introduced a new AI capability called “Computer Use,” which allows AI agents to autonomously perform tasks on computers by mimicking human actions such as using a mouse and keyboard.

Google現開放用戶透過Gemini API試用Gemini 2.5 Computer Use預覽版，以人工智慧代理在網頁上執行點擊、輸入與拖曳等，並內建安全審查與使用者確認機制，阻擋惡意使用與意外行為等操作 ...

Microsoft has announced a feature called ' computer use ' that allows AI agents to click buttons, select menus, and type text into input fields on websites and apps. Copilot Studio, a tool for ...

Artificial intelligence company Anthropic announced the launch of a new version of its Claude AI model alongside a developer’s beta for “computer use” on Oct. 22. The computer use beta will allow ...

This week, Anthropic rolled out updates to its AI models and introduced a very interesting Computer Use API that allows AI to control your computer. Imagine a world where your computer anticipates ...

Craig is a reporter for Pocket-lint based in Toronto. He earned a diploma in journalism from Seneca Polytechnic and holds a Media Foundations certificate from Humber College. Craig previously interned ...