Topic · 主题全部主题 →

Microsoft / Copilot

微软的 AI 布局:Copilot 全家桶、Azure AI 基础设施与 OpenAI 合作关系的持续追踪。

1,390条收录
83条精选

精选归档 · 第 3 页

4160 条 · 共 83

6月16日

星期二 · 2 条
12:37
Hacker News 热门(buzzing.cc 中文翻译)精选
AI 评分 76/100
微软旗下GitHub遭遇AI算力短缺,转而向AWS寻求支持

微软旗下GitHub面临AI算力短缺,微软因此转向亚马逊AWS寻求计算资源支持。原文来自Hacker News热门讨论,标题为“Microsoft turns to AWS as GitHub faces AI capacity crunch”。


推荐理由:微软因AI编码需求导致GitHub容量告急,转向竞争对手AWS租用算力,这信号很明确——AI开发工具已从软件功能战升级为超大规模基础设施竞赛,GitHub的可靠性危机可能加速开发者的平台迁移。
04:59

6月14日

星期日 · 1 条
23:54
Satya Nadella@satyanadella精选
AI 评分 65/100
Satya Nadella:没有生态的前沿不稳定http://x.com/i/article/2065582894790365184A frontier without an ecosystem is not stableI’ve been thinking a lot about the future of the firm in an AI-driven economy.This transition is different than any previous platform shift. In the past, we used digital systems to enhance human capital. This is the first time we can create a real cognitive loop between people and digital systems. That is a mind-bender, because it changes how we even conceptualize work inside an enterprise.What is at stake is not some digital tool or system and its use, but how organizations continue to learn, build IP, differentiate, and thrive in a world where AI models can continuously absorb the expertise of humans and organizations and commoditize it.Every company is going to have to build what I think of as human capital and token capital. Human capital comprises the knowledge, judgment, relationships, ingenuity, and pattern recognition of its people, while token capital is the firm’s AI capability it builds and owns.Importantly, human capital does not become less valuable as token capital grows. It only becomes more valuable! I believe human agency will be the driver of token capital growth. Humans will set ambitious goals, connect dots across domains, build relationships, and recognize patterns that matter most. Without human direction, you have compute running in circles.This means the real opportunity is not in picking the best model but instead in building a learning loop on top of models where human capital and token capital compound. You can offload a task, or even a job, but you can never offload your learning. The future of the firm is the ability to compound that learning across people and AI.This requires a new architectural approach where every business is able to build agentic systems that improve over time, while still retaining control over their IP. A company should be able to switch out a “generalist” model without losing the “company veteran” expertise built into their learning system. This is the key “test” of your control and sovereignty in the era ahead.Companies need to turn their workflows, domain knowledge, and accumulated judgment into AI systems that improve with each use. Private evals should capture whether a model is actually improving against outcomes that matter to the business (not just external benchmarks!). Private reinforcement learning environments should let models grow stronger on real traces from inside the organization. Its knowledge base makes institutional memory queryable and use of tokens more efficient.This loop becomes the new IP of the firm. I think of it as a hill climbing machine. And unlike most assets, it compounds. Every improved workflow generates better training signal, which accelerates the accumulation of tacit knowledge unique to the firm. The companies that build this early will have an advantage that is hard to replicate, regardless of any new individual model capability.The last thing any of us want is a world where every company across every sector is ceding value to a few models that eat everything they see. If all the value is accrued by only a few models, the political economy will simply not tolerate it. There is no societal permission for an AI future that hollows out entire industries.Think about what happened in the first phase of globalization where entire industrial economies were hollowed out by outsourcing. The GDP numbers looked fine on the surface, but the displacement was real and the consequences are still being felt. Let us not bring that dynamic into the AI era, with a small number of AI systems capturing all the economic returns, while entire industries find their knowledge commoditized right out from underneath them.In my view, our priority has to be building a frontier ecosystem, not just a frontier model, so value flows broadly across every company, every industry, and every country. One where every organization can own the learning loop that encodes its institutional knowledge, compounding its human and token capital.This is the ethos I’ve grown up with where platforms enable more value on top than is captured inside, and where every company can continuously innovate and build value of its own.When that happens, companies will create value for themselves and for the economy around them. Employees will see their expertise amplified and their judgment become part of systems that make it replicable and scalable and the benefits accrue to the companies and communities around them.That is how companies drive value for themselves and the broader economy. And it is the stable equilibrium we should build together.微软CEO Satya Nadella认为,AI驱动的平台转变首次实现人与数字系统间的认知循环。企业需同时构建人力资本(知识、判断、关系)与token资本(自有的AI能力),且人力资本不会贬值,反而随token资本增长而增值。真正的机会在于建立人力资本与token资本复合增长的学习循环--企业应能替换通用模型而不丢失已内化的专家知识,通过私有评估和强化学习让模型从内部真实轨迹中持续提升。他警告,若所有价值被少数模型吞噬,将重演全球化空心化悲剧,呼吁构建前沿生态系统,让每家企业、行业和国家拥有自己的学习循环。
推荐理由:Nadella 抛出了一个真问题,当模型能吸收一切知识时,企业的护城河是什么。人力资本与 token 资本的双轮循环框架,比空洞的「AI 转型」更有实操感。

6月13日

星期六 · 1 条
06:29
GitHub Blog精选
AI 评分 61/100
GitHub Copilot CLI 在委托任务上变得更具选择性

GitHub Copilot CLI 通过更好的编排实现了更少的任务交接和更快的进度,且没有新增任何配置选项。


推荐理由:官方博客把子代理从默认操作变成了需要权衡的决策,23% 的工具失败减少和明显的等待时间下降,说明 AI 工具的体验升级不一定要加新按钮,改好调度逻辑一样有用。

6月11日

星期四 · 1 条
00:23
GitHub Blog精选
AI 评分 68/100
通过语言服务器为 GitHub Copilot CLI 提供真正的代码智能

GitHub Copilot CLI 现在可以通过安装和配置 LSP(Language Server Protocol)服务器来替代原始的暴力 grep 或反编译方式,从而获得真正的代码智能。


推荐理由:虽然只是 Copilot CLI 的一个功能指南,但用 LSP 替代 grep 带来的代码理解提升是实打实的,搞 CLI 开发的可以直接抄作业。

6月10日

星期三 · 2 条
23:28
IT之家(RSS)精选
AI 评分 73/100
毕业典礼频现"谈 AI 色变",微软总裁史密斯呼吁行业必须回应公众担忧

近几周多场毕业典礼上,演讲者宣传AI技术时遭学生嘘声。普林斯顿应届毕业生曾否决一款疑似借助AI设计的毕业典礼夹克。微软总裁布拉德·史密斯回应称,行业必须严肃可信地回答问题。史密斯主张AI应增强人而非取代人,认为实用AI渗透经济的速度可能比行业乐观预期更慢。微软今年计划投入约1900亿美元资本支出,主要用于数据中心。微软AI负责人穆斯塔法·苏莱曼修正此前“大多数白领工作18个月内自动化”的说法,表示仅指AI执行单项任务的能力。上月微软CFO艾米·胡德在杜克大学演讲全程未提AI,未遭嘘声。


推荐理由:毕业生直接嘘AI演讲,这比任何用户调研都真实。史密斯呼吁行业回应担忧,但微软一边投1900亿美金一边又说不会取代人,这分裂态度反而说明问题。
00:18
GitHub Blog精选
AI 评分 68/100
GitHub Copilot CLI 推出自定义 AI 智能体,将一次性终端提示转化为可重复工作流

GitHub Copilot CLI 新增自定义 AI 智能体功能,使 CLI 能够理解开发者的技术栈和团队工作流,将一次性终端提示转变为可重复、可审查的流程。


推荐理由:GitHub Copilot CLI 的自定义代理把一次性提示变成可重复工作流,相当于给命令行配了个 AI 副驾驶,做自动化的朋友值得一试。

6月9日

星期二 · 1 条
02:46
Ars Technica:AI(RSS)精选
AI 评分 70/100
数周内第二次:微软软件包被植入凭据窃取器

73个软件包被植入自复制窃密程序,一旦被AI智能体打开就会立即运行。这是数周内第二次发现微软相关软件包被伪装成凭据窃取器。


推荐理由:微软官方仓库第二次被供应链攻击者劫持,这次恶意包专门在 AI 编码代理中触发凭证窃取,蠕虫还能伪造 SLSA 信任链,任何用过受感染包的开发者都应立刻排查。

6月8日

星期一 · 1 条
22:08
The Verge:AI(RSS)精选
AI 评分 79/100
微软AI CEO:超级智能即将到来,但不会取代你的工作

微软AI CEO Mustafa Suleyman在Decoder访谈中表示,超级智能即将到来,但不会导致大规模失业。他透露微软与OpenAI于去年10月签署新合同,巩固合作关系的同时,微软获准独立追求超级智能。微软已组建超级智能团队、训练前沿模型,并于本周Build大会上发布7个全模态新模型。他批评Anthropic将Claude描述为有意识的做法,认为消费者产品需要足够好才能克服公众对AI的负面情绪。


推荐理由:Mustafa 首次系统阐述微软的「自给自足」路线,一边甩开 OpenAI 一边定义「人文主义超级智能」,开发者和产品人都该听听这段博弈论。

6月5日

星期五 · 1 条
19:19
swyx@swyx精选
AI 评分 75/100
微软CEO Satya Nadella最新访谈上线chat is he cookedSatya Nadella 在 Latent Space 发布最新访谈,链接见原文。原推文仅评论"chat is he cooked"。

swyx: @MatthewBerman @saranormous @NoPriorsPod @latentspacepod @satyanadella @Microsoft here! https://www.latent.space/p/satya...


推荐理由:swyx 对 Satya 的一对一访谈,微软 CEO 谈 AI 战略的一手信息远比新闻稿有温度,关心大厂路线的人值得读完原文。

6月3日

星期三 · 8 条
12:02
Satya Nadella@satyanadella精选
AI 评分 77/100
Satya Nadella 谈微软 Build 大会主旨演讲Building a frontier intelligence ecosystem together.Highlights from my keynote at Microsoft Build this morning.与我们共同构建前沿智能生态系统。 今早我在 Microsoft Build 上的主旨演讲亮点。

推荐理由:微软 Build 上的战略更新不算意外,但 Nadella 亲自解读的生态整合思路,对依赖 Azure 和 Copilot 的团队来说是半年内的重要路线图。
07:02
Satya Nadella@satyanadella精选
AI 评分 74/100
微软与高通合作推出Project Solara平台With Project Solara, we are building a new platform purpose-built for agent-first devices.Excited to work with @cristianoamon and @Qualcomm on this!通过Project Solara,我们正在构建一个专为智能体优先设备打造的新平台。 很高兴能与@cristianoamon和@Qualcomm合作!

Cristiano R. Amon: We’re shifting from apps and operating systems to agents, and that changes the device experience end to end. Great conve...


推荐理由:微软和高通联手搞了个 Agent 优先的硬件平台 Project Solara,这标志着 AI 竞赛正式从模型卷到了设备,以后什么是智能终端可能要被重新定义。
03:00
Microsoft:Official Blog(RSS)精选
AI 评分 66/100
微软:企业成功的关键在于围绕AI构建集成智能体平台

微软认为,企业成功的关键在于围绕AI构建一个集成的智能体平台,而非仅拥有强大模型。该平台需遵循三大原则:单一集成系统、安全与治理内建、持续改进。微软正构建一个支持多种模型、以开发者为中心的开放平台,让智能体在软件交付、支持、财务等职能中执行长期工作。


推荐理由:微软把agent平台从散装工具升级成统一系统,这个「企业AI操作系统」的思路让治理和持续进化成为内置能力,而不只是模型和算力。
02:45
The Verge:AI(RSS)精选
AI 评分 78/100
微软首款高级推理AI模型MAI-Thinking-1发布

微软在Build 2026上发布了其首款高级推理AI模型MAI-Thinking-1。该模型被定位为“中等规模”,能在“关键”软件工程基准测试中达到领先模型的水平。微软称其完全从头使用干净数据进行训练,未涉及从第三方模型进行知识蒸馏。这标志着微软在自研AI模型上迈出重要一步,此前其主要依赖OpenAI。近期两家公司已重新协商合作协议,关系有所松绑。

另有 3 家信源报道X:Rohan Paul (@rohanpaul_ai)The Decoder:AI News(RSS)X:Satya Nadella (@satyanadella)
推荐理由:微软自己从头训练的推理模型,不用任何第三方蒸馏数据,这可能是微软系 Agent 和 Copilot 底座更换的信号,值得盯着看。
01:56
GitHub Blog精选
AI 评分 72/100
GitHub Copilot应用:智能体原生的桌面体验

在微软 Build 2026 大会上,GitHub 发布了新的工具和更新,并将 Copilot 应用定位为“智能体原生的桌面体验”。其核心目标是让 AI 智能体能够以用户已经习惯的方式进行工作。

另有 1 家信源报道The Verge:AI(RSS)
推荐理由:GitHub 把 Copilot 从插件变成了独立桌面应用,Agent 不再躲在 IDE 背后,这是 AI 编程从辅助工具走向桌面中枢的标志,开发者现在可以直接在桌面上跟 AI 协作了。

5月26日

星期二 · 1 条
07:58