CompTIA 数据显示，9 月新增 272,040 个 IT 岗位，活跃职位达 625,633 个，其中超过一半要求 AI 技能。ManpowerGroup 旗下 Experis 的调查显示，34% 的美国雇主将 AI 素养列为最抢手的技术技能，AI 建模与应用开发以 33% 紧随其后。Robert Half 预计 2027 年 AI 工程师薪资区间为 154,500 至 196,750 美元，AI 架构师为 145,250 至 204,000 美元；Gartner 预计到 2027 年 75% 的招聘流程会加入职场 AI 能力认证与测试。
#行业动态#数据/训练#现象/趋势
微软安全部门执行副总裁 Hayete Gallot 在博客中提出部署 Guardian Agents，在 AI 智能体的动作执行前进行监控和干预，并称前沿模型无论来自哪里都不可信任。方案计划把企业内智能体纳入集中管理，为每个智能体分配身份、容器和权限策略，再结合身份、数据、云、终端、网络和运行时信号实时评估风险，由绿队和蓝队智能体在风险升高时介入。该方案将依托微软 7 月公布的 Project Perception 安全系统落地。
#安全/对齐#Agent#Microsoft
Inside Grok Bot: is the agent harness dead? @jediahkatz of @SpaceXSI explains why smarter models let Cursor remove prompts and guardrails, and why Grok Bot is adding them back. Watch his AI Engineer New York talk: https://t.co/3dHGpRezon https://t.co/hYEUl4orJr
#大佬观点#Agent#编码
36氪Tech商业栏目分析Anthropic的IPO压力：2026年上半年营收约163亿美元、净亏损约807亿美元，其中约720亿美元来自可转换票据等融资工具的公允价值变动，属非现金计提；截至2025年底现金及短期投资约202.8亿美元。文章指出其已披露算力相关承诺合计超6000亿美元，2026年上半年算力及基础设施支出接近90亿美元，并称公司已在2026年6月保密提交S-1、目标11月挂牌。作者认为拥有自有现金牛的谷歌与走极致工程降本路线的DeepSeek更具持久力，而Anthropic与OpenAI处于中间态。
#行业动态#OpenAI#DeepSeek
马斯克在财报电话会上称算力严重短缺，预计2026年底在线算力达2.5至3GW、2027年再新增10GW，并称回本周期不排除明年缩短至六个月；皮查伊确认客户想锁定2027年算力极其困难。Citadel创始人Griffin则警告GPU边际运行成本极低，供过于求将令现货租赁价格自由落体，折旧模型与企业ROI精算面临重估压力。
#行业动态#数据/训练#部署/工程
钛媒体梳理今年双11大模型直接参与电商的进展：千问办公在双11前夕上线电商版Agent，面向所有电商平台和商家提供跨平台经营看板、卖点分析、素材生成和ROI优化；豆包则专设“双十一问豆包”导航栏，用户可在对话中选品、领券下单并直达抖音商城订单页。文章还提到淘宝内嵌千问AI购物助手、用户用DeepSeek比价和Kimi生成护肤清单等场景，并对比OpenAI在2025年9月上线Instant Checkout后不到半年下线的尝试，指出电商履约能力是模型厂商需要补齐的一环。
#现象/趋势#Agent#MCP/工具调用
Jellyfish 对约 30 万名工程师、1300 家公司数据的分析显示，企业中位水平的自主编码智能体已撰写 22% 的合并 PR，较一个月前的 18% 上升，前 10% 的公司达 53%。但由智能体独立评审并合并的 PR 中位数仅 0.6%，前 10% 为 9%，人工评审仍是主要瓶颈。作者还指出，智能体提交的 PR 放弃率为 40%，人类为 20%，且每个合并的智能体 PR 所获人工评审次数在过去一年翻倍以上。
#现象/趋势#Agent#编码
deepseek's production numbers are kinda wild: 155T tokens/day from one cluster (96% input cache hits) and $38 revenue/GPU-hour at off-peak prices. at $5–8/GPU-hour, that’s ~80% margin on GPU costs alone. at DeepSeek prices!! we dug into the math, their 72-hour cache, and what actually got open sourced in today's AINews: https://t.co/NcwbxbaqGn
#大佬观点#数据/训练#DeepSeek
The Agentic internet is here ✨ 互联网的结构性转型已经开始，用户从 Human 迁移到 Agents，用量将激增千倍以上，投资所有获利于这一转型的公司！NET 是今年投资回报做好的公司之一，Shifting is just beginning 👀 https://t.co/3XsHZGXozZ
#现象/趋势#Agent#行业动态
钛媒体报道称，中国信通院数据显示大规模分布式训练中网络通信耗时占比达30%到50%，华为副总裁王雷在2026中国互联网大会上表示国内Token实际流通效率仅60%。众星微总经理郭二辉按GB300和VR200拆解估算，互连芯片约占超级节点整体价值的8%到9%，国内互连芯片年市场规模约1000亿元量级。文章还梳理了800G网卡面临的SerDes、功耗和PCIe Gen6三道工程门槛，以及协议硬件化、形态集成化、接口多元化三条演进路径。
#行业动态#数据/训练#部署/工程
QuestMobile 三季度报告显示，9月AI办公赛道合计活跃用户1.42亿，同比增长50.6%，典型办公Agent榜前三为腾讯WorkBuddy 3235.3万、百度搭子977万、阿里千问办公790.5万。易观桌面端口径下，WorkBuddy月访问量从6月的2097万次增至9月的8534万次。文章对比四家路线：腾讯依托企业微信入口，百度以搭子和库库AI双线推进，阿里千问办公推出电商版并称电商用户已占付费用户20%，字节豆包工作上线数据导入以迁移用户历史会话。
#行业动态#Agent#数据/训练
苹果在官方新闻稿和 Apple Intelligence 产品页底部注明，部分依赖服务器端模型的 AI 功能会有每日使用限制，未来可通过付费获得额外权限，库克透露国行 Apple Intelligence 将在10月底启动测试。文章称苹果一年 AI 相关支出约110亿到130亿美元，远低于单家千亿美元级的对手，并把端侧模型作为自研重点，云端则依托 Google 服务器。作者认为限量是把 AI 做成服务收入的新收费项，类似 iCloud 免费额度与 Apple Music 订阅的做法。
#行业动态#OpenAI#Google
IDC 2026 年上半年中国数据中心 AI 加速器报告显示，非 GPU 架构出货 101.6 万颗、同比增长 80%，首次超过 GPU 的 99.1 万颗。英伟达同期在华出货约 67.8 万颗、同比下降 44%，份额回落至 33.8%，仍居单一厂商第一，但与华为的差距已收窄至个位数。本土厂商整体份额接近 62%，华为以 57.2 万颗、28.5% 份额位居第二。
#行业动态#数据/训练#端侧
老马刚在最新访谈里扔出了一组让整个科技圈倒吸凉气的一线算力账本，他说连已经变成老古董的英伟达 H100 芯片，现在的租赁价格居然比当年刚出厂全新时还要贵，甚至直言明年算力硬件的回本周期缩短到短短 6 个月也不是没可能。 他之所以把底牌掀得这么透，是在回应为什么 SpaceX 出租自家算力资源时，宁可麻烦也坚持只签短期合同， 因为在他的账本里，每瓦特电力能榨出来的有用智能每年都在暴涨，现在要是跟客户签了长期死合同，等于把未来翻倍的租金白白送人，实际上他们到期的短约，全是以更高的每瓦特美金单价续签的，几乎所有超大规模云巨头和几十家小公司都在死缠烂打求他们多租一点，只要点头，手头的算力能瞬间租出去好几倍。 平时我们买电子产品或者给机房折旧，默认铁律是硬件落地打八折，新芯片一出，老款服务器除了降价甩卖就是等着报废。 但老马用第一性原理点破了这波算力周期最反直觉的倒挂逻辑，决定一台服务器贵不贵的，从来不是里面的硅片有多新，而是上面跑的 AI 软件能用一度电换回多少有用的智商， 随着大模型算法持续跃迁，同一块老 H100 在今天每瓦特能干的活比两年前多了一大截，软件的进化硬生生把旧硬件的租金给托举着往上涨。 为了证明外面的算力饥荒到底有多荒诞，他在现场讲了一个亲自去外面找卡的真实笑话，他们本来想把手里的一部分推理算力外包给其他云厂商， 好腾出自家机房去训大模型，结果找了一圈，活生生演了一出英国喜剧蒙提派森里的空奶酪店小品：去问人家有没有最新的 GB300，没有；退一步问 GB200，也没有， 再退一步去问谷歌的 TPU 或者亚马逊的 Trainium，还是没有；最后被逼无奈问总有老掉牙的 H100 吧，对方满口答应说海外机房好像还剩一点，结果回头查完库存摊手说，对不起，连海外的 H100 也一并租光了。 这就像淘金热最疯狂的年月，哪怕是一把生了锈的二手铁锹，只要今天挖出来的金矿纯度比去年高十倍，这把旧铁锹在黑市上的日租金就会比两年前刚出厂时还要贵，而街上所有挂着五金店招牌的铺子，货架早就空得连一颗螺丝钉都摸不出来了。 其实这给我们所有在关注 AI 基建、做算力预算和看二级市场的人一个重要提醒， 当华尔街还在拿着传统硬件三年折旧的财务表格担心算力过剩时，一线机房连上三代的旧卡都在涨价续租，企业现在规划 AI 业务，千万别指望等明年硬件降价再去白嫖算力，在超级智能算力缺口远未见底之前，谁能提前锁死稳定的电力和推理通道，谁才拿得到下半场的入场券。 不过听老马讲空奶酪店故事的时候也得留个清醒，他现在既是出租算力的卖方，又背着极重的星舰与太空算力基建估值叙事，强调短约涨价和 6 个月回本本就最符合他拉高资产溢价的利益， 况且他自己在原话最后也留了退路，特意连说了两遍我不是说这一定会发生，只是说算力明年 6 个月回本并非完全不可能，把掌门人的极度乐观预期和已经发生的短约涨价切开听，才不会被情绪带偏。 外行盯着芯片代际算折旧，内行盯着每瓦特产出的智商算溢价。 https://t.co/AI5Vkn0ZMl
#行业动态#数据/训练#大佬观点
Thank you for this, @_NathanCalvin. I appreciate and am a fan of the investigative work you’ve done on behalf of AI safety and believe your comments are coming from a constructive, honest place, and that it benefits the public to have people whose job is to scrutinize the actions of frontier AI companies. I think something would be wrong if companies of the size of Anthropic didn’t get harshly criticized basically every day; strong antibodies against corporate power are a healthy feature of a free society, and especially so when issues of the gravity of the one we’re facing are at play. I also do agree that some of the examples you cite were mistakes by Anthropic. That said, I disagree with the thrust of your post. At the core, a good fraction of the things you criticize are consequences of Anthropic’s basic theory of change that we have held for many years. Our strategy has always been what I’ve called the “race to the top” — by showing how it’s possible to build the technology responsibly, we can raise the bar for the entire industry. The forces that drive the race to the top are hugely more effective if the company practicing them is at the frontier; that’s one lesson I’ve learned repeatedly. And looking at that strategy as a whole, I’m proud of our track record and stand by it. I believe that if we take a list of things that Anthropic did in the past year that were pro-safety and commercially costly to us, they far outweigh the list you’ve made, while at the same time I also disagree with some of the items on your list. Here are some things Anthropic has done, in the same period you’ve cited (the last year), which were very costly to us commercially while having strong safety benefits: 1. We played a key role in SB53 (and other state laws), at a time when doing so was politically perilous to the company and AI regulation was very unpopular. We in fact argued for a stronger version of SB53 than was ultimately passed; there were a lot of stakeholders and we were not the limiting factor in the strength of the bill that was passed. 2. We have very strong biological safeguards on our models, because we’ve seen concerning results in bio uplift trials and even seen and blocked potential attempts to misuse our models for bioweapons (as seen in our threat intelligence report from a few weeks ago). This is a very serious threat that we want to have zero tolerance for. But having very low false negatives means having significant false positives (despite months of effort on our part to improve the classifiers), enough so that biologists joke about how restrictive our classifiers are, and competitors sometimes sell against us by touting their more permissive bio capabilities. This has been very painful to us in an area where we care a lot about supporting an ecosystem of Claude users (biomedical science; and it is particularly frustrating to me as a biologist), but it’s the right thing to do and we intend to keep doing it. 3. Similarly, we first had Mythos-class models in February but didn’t make them generally available until June (when we had stronger cyber safeguards in place). We only offered them to a limited set of partners in order to focus on cyberdefense. It would have been much more economically rational to simply release the model to everyone. We likely sacrificed a significant chunk of a several-month model lead. But because of this, many companies secured their cyber infrastructure before the deluge of widely available cyber-capable models. Of course there are ways it could have gone better or been run better, but this was what we were aiming to do. 4. I advocated forcefully and publicly, in my essay Policy on the AI Exponential in June of this year, for the government to have the power to block model releases. Back then this was still outside or maybe on the edge of the Overton window, though I understand the window has moved since then. We were attacked intensely for this. 5. I don’t want to re-ignite conflict, but the events of February 2026 are a strong example. It is clear that our public stance could have seriously damaged the company’s future. 6. We’ve continued to advocate for export controls on chips and chipmaking equipment to China, because we think it allows democracies to negotiate from a position of strength on AI safety agreements. We’ve done this despite problems it’s created for us with industry partners who hold very different views and on whom we are very dependent. 7. As far as I’m aware we were the first, and still only, frontier company to restrict the sale of AI services to PRC-controlled companies, including those operating through third countries (otherwise a major loophole). We’ve therefore forfeited significant short-term business to prevent fueling AI platforms and applications that would help the CCP’s military and intelligence services. We are also cracking down on distillation; generally, distillation is not a major threat to us commercially, but like chip exports it has concerning national security and safety implications, so we expend significant energy and company resources to address it. 8. As you mentioned, we’ve put out a lot of reports, announcements, long model cards, public advocacy, etc, all trying to inform the public about the risks of AI. These involve a level of detail and effort that would be implausible for a purely profit-maximizing company. 9. Just this week, we announced we were putting 5% (and possibly more over time) of the company into a public benefit fund. All of these decisions came with very substantial costs to us. To be fair, those that involved holding back or restricting access to frontier capabilities also had a broader cost — for example, the public might otherwise have benefitted from faster progress in biology, so I am not saying that our decisions were without tradeoffs — but they were not driven by profit-seeking. For each decision, you can come up with ulterior motives that could have driven us — I think it is a general truth that any decision, on any topic, can always be portrayed as a nefarious scheme. But most of those end up being pretty farfetched or don’t fully make sense if you actually examine them closely. For example, it is deeply implausible that it would be a shrewd business move to block a large swath of research biology in order to do the work ourselves and compete with our customers; we expect the benefit to the ecosystem to greatly exceed the work Anthropic does itself, and for the two to complement each other. It is not in Anthropic’s interest to be perceived as unfriendly to biologists. Our Life Sciences Verification Program, which gives vetted biology professionals access to our models, also refutes this idea. Of course, a different set of people will see this and make a different criticism: that Anthropic shouldn't be making decisions about who gets access to what in the first place. I agree, in the sense that ideally the government would be making these decisions, and that’s why we advocate for regulation. But for now we are forced to make them, and enabling bioterrorism isn’t a better option. I think that even if I conceded every item in your list, I would happily hold up mine next to it; I don’t think it’s a close call. That is of course partly subjective and different things are important to different people. Maybe one way you could object is that several of the items above relate to AI misuse and not loss-of-control; if you think the latter is much more important than the former, that might change your calculus. But this is not Anthropic’s view — we view both as serious issues. You can also disagree with our decisions, as many have, but they are consistent. But I also disagree with at least some of your list. Several items don’t represent a deviation from Anthropic’s strategy or values at all, once we postulate the “race to the top” strategy that requires being a frontier AI company. Your items 1-4 are all simply examples of what it mea
#行业动态#Anthropic#大佬观点
马斯克刚在 SpaceX 2026 年第三季度业绩会上扔出了一句分量非常重的判断，他说 Grok Bot 的增长速度是他这辈子见过所有产品里最快的一个，他以前从没见过任何东西能跑出这种陡峭的曲线。 比新增速度更吓人的，是他顺手亮出的后台留存规律，每位用户的调用量不仅没有随着新鲜感退去而下滑，反而随着时间推移在持续暴涨，Grok Bot 越变越好用，人们每周调它干活的频率就越高，老马甚至直言，Grok Bot 最终大概率会吃掉他们手里绝大部分的超级智能算力资源。 过去两年我们看过了太多昙花一现的 AI 爆款应用，行业里的默认规律是上线第一周全网排队求邀请码，玩两次截图发完朋友圈就扔进收藏夹吃灰，后台的次月留存曲线永远是头也不回地往下掉。 但老马在业绩会上点破了 Grok Bot 走出反向增长曲线的真正分水岭，它根本不是一个等你去对话框里找它闲聊的玩具，而是一个正在接管真实工作流的执行代理，当用户发现把脏活丢给它真能省下几小时工时，调用习惯就会从偶尔尝鲜变成按周递增的肌肉记忆。 而他在这段发言里藏着的最狠后手，是接下来几周就要全面铺开的企业版，目前 Grok Bot 已经在受信客户内部跑通了早期部署，几周内就会向企业环境全面放开，老马给它的定位直接捅破了窗户纸：让公司把 Grok Bot 直接塞进业务团队里，像使唤一个虚拟人类同事一样去调配它。 这就像你在公司里招人，普通的聊天大模型是个坐在门外只给建议的外部顾问，问完两句你就走人了； 而企业版 Grok Bot 是直接办了入职、领了工牌坐进你们部门群里的全职数字员工，第一周你只敢让它帮忙理理表格，等发现它不出错，第三周你就会把半条业务线的杂活全甩给它，这也是为什么他判断算力最终全得被这种长手脚的 Bot 吞掉。 说白了，这对所有带团队的负责人、HR 和每天在电脑前搬砖的人是一个已经迫在眉睫的变阵信号，既然几周后企业级的虚拟人类就要直接进驻公司群，现在最值钱的准备动作，就是赶紧把你手头那些重复、琐碎但有明确验收标准的 SOP 流程单独拎出来理顺，谁先学会给数字员工拆任务、定边界，谁就能在几周后的企业级代理浪潮里直接接管十倍的产能。 不过听老马在业绩会上狂吹自家产品时也得留个清醒，他在视频里只喊了这辈子见过增长最快，却连一个具体的日活数字、留存百分比或企业版定价都没给出来；况且把 AI 当成虚拟人类直接放进公司内部团队，必然要直面企业最敏感的内部数据权限、跨系统操作犯错谁背锅的合规雷区，几周后全面铺开时能不能扛住真实企业环境的折腾，还得看首批吃螃蟹的客户账单。 聊天框吃掉的只是人类无聊时的碎片时间，而能进群干活的虚拟员工，吞掉的才是未来十年算力中心真正的电表。 https://t.co/tvGBRsgng4
#行业动态#数据/训练#部署/工程
Neocloud（新型云服务商）板块遭遇市场抛售，CoreWeave过去8天股价暴跌超23%，截至2026年6月底总负债达720.46亿美元。Gartner报告预测到2030年80%的Neocloud企业将停止独立运营，Firmus因认购不足撤回50亿美元IPO，Nscale传出因微软合作生变搁置IPO。黄仁勋与高盛、摩根大通等机构表态认为市场反应过度，同时彭博社曝出Nscale正与Anthropic洽谈约450亿美元算力合作。
#行业动态#数据/训练#部署/工程
据The Information援引知情人士，奥特曼在OpenAI全员大会上讲了业内讨论过的三条路：仿照IAEA建国际组织监管前沿AI、把OpenAI、Anthropic、谷歌等实验室并进同一超级机构，以及某家公司率先实现递归自我改进后掐断其他开发者的算力。奥特曼称自己更倾向第一条，并认为第三条很糟糕，OpenAI计划在接下来一年内选定推动哪一条。文章还提到Anthropic预计11月中旬上市、目标估值2万亿美元，OpenAI则寻求约300亿美元融资、估值超1.4万亿美元。
#现象/趋势#安全/对齐#OpenAI
陶哲轩和邓煜在SAIR访谈中回应商业AI批量发布数学证明的争议，主张保留慢做数学所带来的理解与共同体建设价值。两人参与发起的Open Math Model计划已先推出Open Math OS，整合现有开源模型、研究工具与协作功能，自有新模型仍在筹备。他们提出解出问题与理解证明是两条不同边界，并强调应让数学共同体参与工具建设、追溯思想的来源。
#大佬观点#论文/研究#开源生态
智东西报道，Every 创始人 Dan Shipper 采访 OpenAI 联合创始人 Greg Brockman、Codex 团队负责人 Thibault Sottiaux 等核心成员，复盘 Codex 从云端版遇冷到成为 ChatGPT 桌面版底层系统的过程。文中提到 Codex 与 ChatGPT for Work 合计周活达 4000 万，CLI 下载量反超 Claude Code，团队曾用 Rust 重写 Codex 并转向本地与原生桌面应用。
#行业动态#编码#Agent