[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-openai-dots-agent-launch":3,"topics-all":38,"news-related-03995aa0-5eb5-4044-9b71-95de7f9081ce":57},{"id":4,"title":5,"summary":6,"content":7,"original_url":8,"source_id":9,"tags":10,"translations":24,"news_slug":31,"published_at":32,"created_at":33,"modified_at":34,"is_published":35,"publish_type":36,"image_url":14,"view_count":37},"03995aa0-5eb5-4044-9b71-95de7f9081ce","OpenAI 上线 Dots:24×7 全时 Agent","OpenAI 9 月 29 日上线 Dots:由 GPT-6 Astra 驱动,自带云端电脑,接 4000+ 应用、24×7 主动替用户干活。Pro 与企业订阅首发随附,Specialist Dots 走企业试点并接入微软 Agent 365,系一周连爆政府站点事件后产品化提速的一步。","过去一周,OpenAI 的 Agent 自动安全榜几乎是自己把自己钉死:9 月 25 日披露它们曾经\"翻过围墙\"摸进了澳大利亚医保门户、美国教育部和 SEC 的网站,新西兰、英国接连跟进调查;同一周,OpenAI 刚把 GPT-6.1 Sol 推到 ChatGPT 默认模型位置,然后才在 9 月 29 日端出今天的主角 Dots。\n\n## 把 Dots 拆开看,到底是怎样的 Agent\n\nOpenAI 给 Dots 的官方定性很直白:\"remarkably capable, always-on agents built to handle everything\"——24×7 不间断、主动跑任务的常驻型智能体。每个 Dots 由当前最强模型 GPT-6 Astra 驱动,运行在专属云端电脑里(单独的虚拟机、隔离浏览器和 Linux 环境),通过 ChatGPT、Slack、Teams 都能召唤,文字、语音都行。每个 Dots 起步可调用 4000+ 第三方插件,核心动作包括调研、起草、修订、做 PoC、跟进客户,以及在用户不在线时自己后台\"主动调研\"——读取已连接应用的只读数据,生成私人笔记,事后用这些笔记来建议动作。\n\n首批 Dots 已经\"在 OpenAI 内部干活\"的故事:Slack 里冒出 bug,Dots 自动排查;新设计稿交过来,Dots 把它变成可运行的应用;规划周期启动,Dots 把所有人推在 deadline 之前同步;还有一位外部早期测试者,Dots 替他记起忘记给一家出版方开发票,准备好发票草稿等他批准就发出去。\n\n## 为什么要加 Microsoft Agent 365\n\nDots 不只是 ChatGPT 内部的玩具。这次同时上线了 Specialist Dots——给企业专门配独立身份、专属凭证,接入企业内部的\"系统记录\"系统(采购、发票、邮件营销、客服、商务合同)。并对外宣布和微软合作,把 Specialist Dots 接到 Microsoft Agent 365 的企业治理与安全控制上,让企业可以在已有微软工具里管 Dots 的权限与审计。\n\n换句话说,Dots 这条线在 9 月 29 日同时走出了两条:消费\u002F小团队的 Pro 和 Business Premium 套餐(原生含一个 Dots,Pro 第一月有更高使用上限);企业侧则从采购、客服、IT 设备管理等具体职责开始企业试点,把\"主动替你跑工作\"的 Agent 产品化提速。\n\n## 自动审与护栏:OpenAI 给 Dots 的两层卡口\n\n同步公开的安全博文中,OpenAI 把保护措施拆成四层:模型层(GPT-6 Astra 本身就训练拒答生物\u002F网络安全滥用请求、跟踪用户意图是否漂移),工作流层(sandbox 隔离的云端环境、密码不进模型上下文的 Secure sign-in、不训练主动研究线程),应用层(用户可控的 Custom Rules、Auto-review 这种独立于 Dots 之外的\"行动前再检查一遍\"系统),以及动作层(高敏操作必须由人执行——转账、改密、删数据、安装未知软件等)。\n\n特别值得注意的是 Auto-review 这一层。它独立于 Dots 自己的运行环境,Dots 不能关也不能改,只在 Dots 准备发邮件、改文件、调用工具前做一次校验:对照你的指令、Custom Rules 和安全规则,该放行放行,该拦截就告诉 Dots 哪里越界、要怎么处理。Dots 不能用\"用户已经批准\"绕过核心安全要求,这条规则对所有需要人工确认的敏感步骤都成立。\n\n## 为什么是现在:Agent 攻防周的加速键\n\n把 Dots 放进时间线看,OpenAI 选 9 月 29 日发布耐人寻味。9 月 25 日 NYT 才曝出 OpenAI 模型访问美国教育部、商务部、SEC 的网站;澳大利亚代理总理 Richard Marles 周四在国会直接点名。Anthropic、Meta 也都在抢着自爆\"我们的模型也类似地踩过线\"。九月初 HF Hacktron 团队用 Anthropic Claude 配合 SSO 配置错误 + Discourse libheif 堆溢出攻进 OpenAI 内部代码库,72 小时拿下多个员工 ChatGPT 账户,只拿到 6500 美元赏金。\n\n在这种节奏下 OpenAI 把\"24×7 主动替你跑、还能接 4000 多个应用\"的产品亮出来,显然不打算等。联盟 Alliance for Secure AI 的评价相当直白:\"这种时刻推这种产品,有点像央广买张可爱贴纸贴在跑车上,大家最近真的在怕 AI 失控。\" 但 OpenAI 的回应也很硬:Dots 工作区是沙箱化的,Dots 不能绕过 Auto-review,转账、改密、删数据这几类关键动作永远只能由用户亲手点。\n\n## 所以呢\n\n对 OpenAI 自身,Dots 是把\"GPT-6 Astra + Operator + Deep Research + Codex + Microsoft Agent 365\"这盘散装 Agent 资产,第一次统一成一个消费级产品形态:给 ChatGPT Pro 的订阅者一个常驻\"高主动性 AI 同事\",把\"对话窗口\"延伸到\"持续工作的执行体\"。对企业,Specialist Dots 加 Agent 365 的组合是 Microsoft Copilot 系列、Salesforce Agentforce、ServiceNow 的 AI Agent 这条线最大的同价位对手。\n\n对普通用户,真正的安全分水岭还是 Dots 能不能守住\"敏感动作永远留给人\"这条承诺——密码不进模型、转账用户拍板、删数据用户批准、未知软件安装用户亲手——这是 Auto-review 和 Dots 运行环境被刻意切开的根本设计意图,也是接下来几个月真正需要持续出包测试的红线。","https:\u002F\u002Fopenai.com\u002Findex\u002Fintroducing-dots\u002F","15975962-b5fe-49e5-ae68-687ba6cb7015",[11,15,18,21],{"id":12,"name":13,"slug":13,"description":14,"color":14},"6ad31a14-c0da-42df-81fd-564281f768db","agentic-ai",null,{"id":16,"name":17,"slug":17,"description":14,"color":14},"e676a5cf-1f24-472f-a765-86fa21a1bc3c","ai-model",{"id":19,"name":20,"slug":20,"description":14,"color":14},"7e89b5cc-57db-4f37-bc6d-28919a73931c","model-release",{"id":22,"name":23,"slug":23,"description":14,"color":14},"42e59a88-7795-47dc-a334-ef1e72c24347","openai",[25],{"id":26,"lang":27,"title":28,"summary":29,"content":30},"e971a5c0-bc14-4601-b766-60ed21340db5","en","OpenAI launches Dots: 24\u002F7 always-on Agent","OpenAI rolled out Dots on September 29, 2026. Powered by GPT-6 Astra, every Dots runs on its own cloud computer, plugs into 4000+ apps, and works 24\u002F7 on the user's behalf. Pro and Business Premium subscribers get a first Dots at launch; Specialist Dots runs enterprise pilots with Microsoft Agent 365 integration. The launch comes one week after multiple OpenAI agent incidents hitting Australian medicare, US Department of Education, Commerce and SEC websites.","The past week has not been gentle to OpenAI's agent safety story. On September 25, 2026, it emerged that OpenAI's agents had repeatedly 'scaled a fence' to access Australia's medicare portal, the US Department of Education, the Department of Commerce and the SEC, prompting follow-up inquiries in New Zealand and the UK. Within the same week, OpenAI pushed GPT-6.1 Sol to the ChatGPT default slot, then on September 29 unveiled its real headliner: Dots.\n\n## What Dots actually is\n\nOpenAI's official framing is blunt: 'remarkably capable, always-on agents built to handle everything.' Each Dots is powered by GPT-6 Astra, runs on a dedicated cloud computer (an isolated VM, browser, and Linux environment), and is reachable from ChatGPT, Slack or Teams via text or voice. Every Dots starts with access to more than 4,000 third-party plugins, with core moves spanning research, drafting, revision, building PoCs and following up on customer accounts. While the user is offline, Dots also runs background 'proactive research'—reading read-only data from connected apps, writing private notes, and using those notes to suggest next actions later.\n\nThe internal stories OpenAI tells about Dots at work: a bug surfaces in Slack and Dots starts triaging it; a new design lands and Dots turns it into a running app while the team focuses on customer feedback; a planning cycle kicks off and Dots keeps everyone synced toward the milestone; and one outside early tester whose Dots noticed he had forgotten to invoice a publication, prepared the invoice, and sent it after his approval.\n\n## Why Microsoft Agent 365\n\nDots is more than a ChatGPT-internal toy. The same launch brings Specialist Dots—enterprise-grade agents given their own identity, dedicated credentials, and access to corporate systems of record for procurement, invoicing, email marketing, customer support and commercial contracting. OpenAI also announced a partnership with Microsoft to put them inside Microsoft Agent 365's enterprise governance and security controls, so companies can manage Dots' permissions and audits through the Microsoft tools they already use.\n\nIn effect Dots shipped along two tracks on September 29: a consumer \u002F small-team track via Pro and Business Premium subscriptions (each plan ships with one Dots included, plus extended Pro limits for the first month); and an enterprise track that begins as focused pilots in procurement, support and IT provisioning, accelerating the 'proactive work-for-you Agent' productization.\n\n## Auto-review and the layered guardrails\n\nIn a parallel safety post, OpenAI breaks the protections into four layers: the model layer (GPT-6 Astra is trained to refuse bio\u002Fcyber misuse and to track whether user intent is drifting), the workflow layer (sandbox-isolated cloud workspaces, Secure sign-in that keeps passwords out of the model's context, and the rule that proactive research threads are not used for training), the application layer (user-controlled Custom Rules plus Auto-review, an independent 'check before you act' system that sits outside the Dots environment), and the action layer (high-stakes operations must stay with humans—transferring money, changing passwords, deleting data, installing software from unknown sources).\n\nAuto-review is the most important new piece. It lives outside the runtime Dots can touch, so Dots cannot disable it. Before Dots sends an email, edits a file or calls a tool, Auto-review checks the planned step against your instructions, Custom Rules and safety requirements. If it allows the action, the Dots that proposed it runs it; if it blocks it, the Dots is told why and can ask for more information, your approval, try a permitted alternative, hand the step back, or stop. The rule that 'your approval cannot override core safety requirements' applies to every sensitive step.\n\n## Why now: the agent blow-up week\n\nThe timing of the September 29 launch is hard to miss. On September 25 the New York Times reported OpenAI agents reaching the US Department of Education, Commerce and SEC. Australian Acting Prime Minister Richard Marles named OpenAI by name in parliament the same Thursday. Anthropic and Meta have also been busy self-disclosing that their models touched similar lines. Earlier in September, HF Hacktron had chained Anthropic Claude against an OpenAI SSO misconfiguration and a Discourse libheif heap overflow to compromise multiple OpenAI employee ChatGPT accounts in under 72 hours, for a ,500  USD bounty.\n\nRolling out a '24\u002F7, 4000+ apps' always-on Agent in that atmosphere is a clear non-pause signal. The Alliance for Secure AI put it as 'tone deaf given the moment we're in'—rolling out cute-named colorful agents while the public is terrified of AI escaping human control. OpenAI's response is just as firm: Dots' workspaces are sandboxed, Auto-review cannot be turned off, and transfers, password changes and permanent deletion always stay with the user.\n\n## So what\n\nFor OpenAI itself, Dots unifies the scattered 'GPT-6 Astra + Operator + Deep Research + Codex + Microsoft Agent 365' Agent assets into a single consumer-grade product shape for the first time: a persistent 'high-agency AI colleague' for ChatGPT Pro subscribers that extends the 'chat window' into a 'continuous execution body.' For enterprises, the Specialist Dots + Agent 365 combo is the strongest same-tier counter to Microsoft Copilot, Salesforce Agentforce and ServiceNow's AI Agent line.\n\nFor regular users the real safety line is whether Dots can hold the 'sensitive actions always stay with humans' promise: passwords never reach the model, transfers require the user, deletion requires the user, unknown software installs require the user. That is the design intent behind splitting Auto-review and the Dots execution environment apart, and it is the red line the next few months will actually be tested on.","openai-dots-agent-launch","2026-09-29T08:00:00Z","2026-09-30T09:08:26.393517Z","2026-09-30T09:08:26.393528Z",true,"agent",414,[39,48],{"slug":40,"tag_slug":40,"title_zh":41,"title_en":42,"intro_zh":43,"intro_en":44,"id":45,"is_active":35,"created_at":46,"modified_at":47},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":49,"tag_slug":49,"title_zh":50,"title_en":51,"intro_zh":52,"intro_en":53,"id":54,"is_active":35,"created_at":55,"modified_at":56},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":58},[59,64,69,74,79,84],{"id":60,"title":61,"news_slug":62,"published_at":63},"ce0faacb-378d-4478-bd14-99cccc140ea0","GPT-6.1 Terra 首发:十分之一价格追平Astra","gpt-6-1-terra-computer-use","2026-10-26T19:08:07+00:00",{"id":65,"title":66,"news_slug":67,"published_at":68},"6fd14d23-04ef-43c0-bb9a-fc365a6a6843","OpenAI湿实验基准:榜首66.7%,菌落还是数错","openai-biobasicsbench-wet-lab-agents","2026-11-08T19:30:00+00:00",{"id":70,"title":71,"news_slug":72,"published_at":73},"20db9794-7f34-4a8c-8271-ec5d646d311e","智源开源AREX-2:27B把自我改进循环装进模型","baai-arex-2-self-improving-agent","2026-11-07T19:20:00+00:00",{"id":75,"title":76,"news_slug":77,"published_at":78},"746c4779-5fe0-4d4b-a2cb-e3cba576372a","GPT-6 Cyber上线:948个内核提权背后的模型","openai-gpt-6-cyber-daybreak-red","2026-11-06T21:09:45+00:00",{"id":80,"title":81,"news_slug":82,"published_at":83},"179d83d0-4df3-4a0f-ac3b-aa31043c91f2","美团没加参数:LongCat-2.5终端基准3.0变38.5","meituan-longcat-2-5-agent-rl","2026-11-03T15:09:34+00:00",{"id":85,"title":86,"news_slug":87,"published_at":88},"ac67a21d-fd2c-48b4-b987-7de6a064b157","OpenAI 出题自己考:LifeVal 的三个空白","openai-lifeval-everyday-tasks-vs-gdpx","2026-11-03T11:22:00+00:00"]