從 GPT-3 到 Gemini 3 的三年：從聊天機器人邁向 AI Agent 時代★ 80
One Useful Thing (Mollick)208 days agoOpinionWharton School professor Ethan Mollick, writing on his blog "One Useful Thing," offers a sharp retrospective on the most fundamental paradigm shift in AI over…
Google DeepMind 發表全新一代 Gemini 3：開啟主動式 AI 與超強推理的全新智能時代★ 98
Google DeepMind Blog208 days agoReleaseGoogle DeepMind officially unveiled its latest flagship AI model — Gemini 3 — in November 2025. This marks a new milestone for Google in the field of…
Google DeepMind 推出 SIMA 2：由 Gemini 驅動、能在 3D 虛擬世界中與你一同遊玩、推理與學習的 AI 代理★ 85
Google DeepMind Blog213 days agoReleaseGoogle DeepMind has officially introduced SIMA 2 (Scalable Instructable Multiworld Agent 2). Compared to its predecessor, the most significant transformation…
給你的 AI 一場面試：如何評估與測試 AI 的真實工作能力★ 80
One Useful Thing (Mollick)214 days agoOpinionAs AI tools (such as ChatGPT, Claude, and others) become more prevalent in the workplace, we are increasingly relying on them for decision-making advice…
AI 如何為北愛爾蘭教師贏回時間：Google Gemini 試點計畫每週幫教師省下 10 小時
Google DeepMind Blog216 days agoReleaseThis official blog post from Google DeepMind details how AI is making a tangible impact in educational settings and helping to reduce the burden on teachers…
Gemini 2.5 Flash-Lite 正式推出穩定版，支援大規模生產環境部署★ 75
Google DeepMind Blog232 days agoReleaseGoogle DeepMind today announced that Gemini 2.5 Flash-Lite — its lightweight AI model that had previously been in preview — has officially transitioned to a…
Google DeepMind 推出實驗性 AI 工具「Backstory」，幫助使用者探索網路圖片的背景與來源★ 75
Google DeepMind Blog233 days agoNew ToolGoogle DeepMind recently unveiled a new experimental AI tool called "Backstory," designed to help internet users deeply explore and understand the background…
搭載 Deep Think 的進階版 Gemini 正式在國際奧林匹亞數學競賽中達到金牌標準★ 90
Google DeepMind Blog233 days agoReleaseThe International Mathematical Olympiad (IMO) has been held annually since 1959 and is the most prestigious and difficult mathematics competition for high…
Google DeepMind 強化其「前沿安全框架」(Frontier Safety Framework)，以應對先進 AI 模型的嚴重風險★ 75
Google DeepMind Blog233 days agoReleaseGoogle DeepMind has recently announced the strengthening of its Frontier Safety Framework (FSF) — a systematic mechanism designed to proactively identify…
Gemini Robotics 1.5 發表：將 AI Agent 帶入實體世界，賦能機器人感知、規劃與工具操作★ 85
Google DeepMind Blog233 days agoReleaseGoogle DeepMind has officially announced the launch of Gemini Robotics 1.5, marking the formal entry of AI Agent technology into the physical world and…
Google 於 Gemini 應用程式推出 Deep Think 功能，並向數學家開放 IMO 級 Gemini 2.5 完整版★ 85
Google DeepMind Blog234 days agoReleaseGoogle DeepMind has officially announced the rollout of the "Deep Think" feature for Google AI Ultra subscribers within the Gemini application. This new…
Gemini 應用程式迎來重大升級：原生圖片編輯功能全面進化★ 75
Google DeepMind Blog234 days agoReleaseGoogle DeepMind has announced a major feature update for the Gemini application: a comprehensive upgrade to its native image editing capabilities. This update…
Google 推出 Gemini 2.5 Computer Use 模型：基於 Gemini 2.5 Pro，支援 API 預覽★ 85
Google DeepMind Blog234 days agoReleaseGoogle DeepMind has officially launched the new dedicated "Gemini 2.5 Computer Use" model, which is now available in preview via API. This model is built on…
2025 年末 AI 實用指南：Ethan Mollick 的主觀使用建議★ 85
One Useful Thing (Mollick)238 days agoTutorialWharton School professor Ethan Mollick has put together a highly personal and practical operating guide for the AI landscape of late 2025. He emphasizes that…
與魔法師共事：在參差不齊的技術前沿驗證 AI 的魔力★ 85
One Useful Thing (Mollick)276 days agoOpinionUniversity of Pennsylvania Wharton School professor Ethan Mollick, in his latest article, compares the experience of collaborating with generative AI (such as…
在 Vercel 上部署：迎來新一代 AI 輔助軟體開發浪潮
Vercel Changelog317 days agoCommentaryAgainst the backdrop of rapid AI adoption, the definition of software and the development process are undergoing fundamental transformation. Through the theme…
回到未來：Hugging Face 推出 FutureBench 評估 AI Agent 的未來事件預測能力★ 75
Hugging Face Blog332 days agoRelease### What is FutureBench? As large language models (LLMs) and AI agents have rapidly advanced, traditional static benchmarks (such as MMLU and GSM8K) face a…
Vercel 推出「AI Cloud」：專為 AI 工作負載打造的統一平台★ 85
Vercel Changelog339 days agoReleaseVercel has officially launched "AI Cloud," a unified platform designed specifically for AI workloads. This marks a major transformation for Vercel from a…
Google DeepMind 推出 Gemini Robotics On-Device：為本地機器人設備帶來高效能、具備通用靈巧性與快速適應能力的 AI 模型★ 85
Google DeepMind Blog355 days agoReleaseGoogle DeepMind has released the "Gemini Robotics On-Device" model, a significant breakthrough that brings advanced Gemini AI capabilities directly to local…
立即上手 AI：實用快速指南 (Ethan Mollick 著)★ 85
One Useful Thing (Mollick)356 days agoTutorialUniversity of Pennsylvania Wharton School professor Ethan Mollick recently published an extremely practical AI quick guide, "Using AI Right Now: A Quick…
Google DeepMind 擴展 Gemini 2.5 家族：Flash 與 Pro 正式商用，並推出全新超低成本 2.5 Flash-Lite★ 85
Google DeepMind Blog362 days agoReleaseGoogle DeepMind today announced a major advancement for the Gemini 2.5 model family. First, the previously preview-stage Gemini 2.5 Flash and Gemini 2.5 Pro…
Google 發表 Gemini 2.5 更新：Gemini 2.5 Pro 進入穩定版、Flash 全面開放、全新 Flash-Lite 推出預覽★ 85
Google DeepMind Blog362 days agoReleaseGoogle DeepMind today announced a major update to the Gemini 2.5 thinking models family, aimed at improving overall performance and accuracy while providing…
Gemini 2.5 迎來全新突破：先進的語音對話與音訊生成功能★ 85
Google DeepMind Blog376 days agoReleaseGoogle DeepMind has announced that its latest-generation model, Gemini 2.5, has achieved new breakthroughs in AI-driven audio dialog and audio generation. This…
Fuel your creativity with new generative media models and tools★ 78
Google DeepMind Blog390 days agoNew ToolGoogle announced new generative media models and tools at I/O 2025, led by Veo 3 for video, Imagen 4 for images, and Flow for AI filmmaking. Veo 3 adds audio generation, while Imagen 4 improves detail, typography, aspect ratios, and up to 2K output. Google also expanded Lyria 2 and Lyria RealTime access, while continuing SynthID watermarking and launching SynthID Detector.
Gemini 2.5 迎來重大更新：Pro 版推出實驗性「Deep Think」深度思考模式，Flash 版性能再提升★ 85
Google DeepMind Blog390 days agoReleaseGoogle DeepMind today announced important updates to its flagship model series, Gemini 2.5. The most noteworthy highlight of this update is a brand-new…
Google DeepMind 揭示通用 AI 助手願景：將 Gemini 擴展為具備規劃與模擬能力的「世界模型」★ 85
Google DeepMind Blog390 days agoOpinionGoogle DeepMind recently published its latest vision for building a "Universal AI Assistant." In this blueprint, the core technical evolution lies in extending…
Google 推出 SynthID Detector 全新入口網站，協助識別 AI 生成內容★ 80
Google DeepMind Blog390 days agoNew ToolAt Google I/O 2025, Google DeepMind announced the launch of the new "SynthID Detector" portal. This tool is designed to address the increasingly serious…
Google 發表 Gemini 2.5 Flash：主宰性價比邊界，首創可精確控制的「思考預算」★ 85
TLDR AI (Buttondown)422 days agoReleaseGoogle has officially released its new model Gemini 2.5 Flash, marking Google's comprehensive dominance over the cost-efficiency Pareto frontier on LMArena…
SOTA 影片生成新里程：Google Veo 2 與快手 Kling 2 正式對開發者全面開放 (GA)★ 85
TLDR AI (Buttondown)424 days agoReleaseAI video generation has reached a major milestone: Google's Veo 2 and Kuaishou's Kling 2, currently ranked at the top of the Artificial Analysis Video Arena…
Google 發表 Agent2Agent (A2A) 協定：攜手 Anthropic MCP 打造跨 Agent 協同新標準★ 85
TLDR AI (Buttondown)430 days agoReleaseAt the 2025 Google Cloud Next conference, Google dropped two bombshells regarding the AI Agent ecosystem. The CEOs of Google and Google DeepMind jointly…