Claude 将能够自动连接那些接受 Claude 作为可信签发方的插件。 更少的重新授权让智能体的使用效率大大提高。 现在我们需要 Duet 广泛发布 👀
搜索“Anthropic”
🚨 AI News | TestingCatalog@testingcatalogAI 评分4242
Alex Albert@alexalbert__AI 评分4242Ultra 比 Max 20x 听起来更顺耳。另外 Ultra 订阅用户不再有 5 小时速率限制。
引用Claude@claudeaiWe’re renaming our Max plans: Max 5x → Max Max 20x → Ultra Max is still $100/month, and Ultra is still $200. If you’re a subscriber, there’s nothing you need to do. We’re also removing the 5-hour limit for verified Ultra accounts, starting today with most existing subscribers. Newer accounts keep a temporary limit until they’re verified, which helps us prevent fraud and abuse. You'll get an email when it's removed. Weekly limits remain unchanged. View our plans here: http://claude.com/pricing
Alex Albert@alexalbert__AI 评分4242关于该计划的历史与现状简述 https://x.com/mboverell/status/2116221457252659275?a=Alj9su7j0fAQE6dVWFA&s=20
引用mike overell@mboverellI lead product for Claude subscriptions. Today we're renaming our Max plans, and I want to explain why. When we launched our top plan in April 2025, it had a single limit: 5 hours. It gave you 20x Pro's limit, so we called it Max 20x. When we added weekly limits to Max 20x, we set them to 10x Pro's (the 5-hour limit stayed at 20x). I get that some new users assumed 20x applied to both limits. We could have done better here, and I'm sorry. The Ultra plan keeps the weekly limit (10x Pro), but we're removing the 5-hour limit so you can keep building without getting blocked mid-session. This starts today and rolls out in stages. To help prevent fraud and abuse, a temporary limit applies until we've verified your account. You'll get an email when your limit is removed. We've got more cooking and can't wait to share it. Thanks for building with Claude 🙏
Simon Willison@simonwAI 评分4242
The Rundown AIAI 评分4242 Sam Altman 谈 AI 安全竞赛的三种终局
Sam Altman 在内部讲话中向员工提出 AI 安全的三种终局方案:由类似国际原子能机构的全球监管者统一管理、将顶级实验室合并为一家服务伙伴国家的机构,或由一家实验室凭借自我改进的 AI 领先并"熔化其他所有人的 GPU",他称第三种"很糟糕"。他倾向第一种方案,并表示需在一年内选定路径,即"pacing 2.0",其中必须包含针对自我改进 AI 的"限速"或停止点。

Simon WillisonAI 评分4242 Addy Osmani 等谈 Claude Code 团队的代码审查方式
Addy Osmani、Morgan Westlee Lunt 和 Qing Wang 在《How the Claude Code team does code review》中提出,"人是否读过代码"只是判断变更风险的粗略替代指标,更该问哪项检查能抓住问题。他们称代码审查已从 PR 末尾的关卡变成持续进行,多数变更在人工查看前已被 Claude 至少审查并修复过一次。
Riley Goodside@goodsideAI 评分4242
Alex Albert@alexalbert__AI 评分4242引用Claude@claudeaiWe’re introducing Claude Fable 5.1 and Claude Mythos 5.1. They're the world’s most advanced models for coding and knowledge work.
Andrej Karpathy@karpathyAI 评分4242
Andrew Curran@AndrewCurran_AI 评分4040AI 2027 已更新。用 Astra 时我们落后曲线五周,而用那个神秘模型 4,我们现在领先六周。
引用Chubby♨️@kimmonismusThe AI 2027 tracker was updated today. And according to the tracker’s estimates, based on AECI scores from METR’s report on Anthropic’s internal models, some of those models are even ahead of the AI 2027 curve.
elvis@omarsar0AI 评分4040
引用Anthropic@AnthropicAIMemory is what allows Claude to learn from experience. In a new Engineering Blog post, we share how “dreaming” helps Claude organize what it’s learned, correct its mistakes, and find what it needs more easily the next time. https://www.anthropic.com/research/components-of-agent-memory
歸藏(guizang.ai)@op7418AI 评分4040Claude 已经能在 CarPlay 上使用了 而且 CarPlay 上可以直接用语音跟 Claude 交流,这个很方便呀。 什么时候 Codex 支持一下?
引用Claude@claudeaiOn iPhone, voice mode now works with Apple CarPlay. Ask Claude something on the drive and keep your hands on the wheel.
Lisan al Gaib@scaling01AI 评分4040
引用Cognition@cognitionOn our price vs performance evaluation, models cluster into several groups. ▪️ Opus 5.5 and GPT-6 Astra lead in intelligence, at average costs of $12.8 and $6.67, respectively. ▪️ GPT-6.1 Sol leads in cost efficiency, scoring 39% at $1.03 per task. ▪️ GPT-6 Luna and Haiku 5.5 lead on overall cost, scoring above 20% at less than $0.50 per task. Chinese OSS models lag far behind, with Kimi K3 and GLM 5.3 scoring at 17%. American OSS models like Beam and Inkling are even further behind with a score of 3%.
宝玉@doteyAI 评分4040引用Techartist@techartist_Nine more lessons from my 3D book about energy, now covering the whole GCSE energy topic: kinetic, gravitational and elastic energy, heating, power, efficiency, energy resources and conservation of energy. Built with Opus 5.5 and Three.js. The full book ↓
歸藏(guizang.ai)@op7418AI 评分4040有人爆料说,Claude Code 的作者 Thariq,离职了。 他是一直以来代表 Anthropic 跟大家沟通的人,他在 Anthropic 的地位类似于 Tibo。这是一个非常重要的信号
引用Arfur Grok@ArfurGrok👀 Thariq Shihipar (@trq212) has left Anthropic.
Ethan Mollick@emollickAI 评分4040“提示注入”一词的发明者怀疑(我也这么认为),这个问题在新模型中可能没那么严重。这与我们的测试结果一致:如今你无法再简单地通过提示注入来让论文拿到更高分数。
引用Simon Willison@simonw@nateberkopec I'm cautiously starting to suspect it may not be a problem with today's strongest models - Anthropic in particular are very confident about their ability to avoid prompt injections - but it's still a real concern for cheaper, weaker models
Aran Komatsuzaki@arankomatsuzakiAI 评分3838引用Nace AI@NaceAIIntroducing Drex Router. A small model that picks the right LLM/SLM and reasoning effort for every step your agent takes. Accuracy: Drex Router 87.6% vs. Jev Router’s 86.4% Try it: https://console.nace.ai COMMENT DREX and get extra $5 in API credits
🚨 AI News | TestingCatalog@testingcatalogAI 评分3838
引用ClaudeDevs@ClaudeDevsWe've made some big updates to Projects in Claude Code! In a project, Claude coordinates your tasks across parallel threads. The coordinator is now much more proactive. It answers your questions in the project chat and gives you updates on the status of your ongoing threads.
宝玉@doteyAI 评分3838@MyDaebakCafe 原帖 https://x.com/MyDaebakCafe/status/2121031084653637728
引用Daebak 앤디@MyDaebakCafeI am actually disgusted by the timing of this drop. And this will get a bit more technical. Let’s do a real technical teardown of this "1,385-sigma Blackwell weight fingerprint" report that OpenAI and Anthropic are breathlessly promoting today, because banking on nobody reading past the headline is how DC narratives get manufactured. First, look at the actual CUDA maths: when PyTorch initialises model weights (torch.nn.init) with a fixed PRNG seed, curand/Philox parallelises random number generation across the GPU’s Streaming Multiprocessors (SMs). Because Hopper has 132 SMs and Blackwell has 148 or 152 SMs, the exact same seed lays out initial weights in a different spatial pattern depending on the GPU’s SM count. In huge MoE models, certain low-gradient null-spaces or unrouted slices preserve faint directional traces of those step-0 weights even after training. So yes, brute-forcing the seed and SM grid can show that a model’s step_0 weight initialisation script ran on a 148/152-SM Blackwell GPU. Now read what the summary explicitly admits in the fine print: 1. Initialisation is not full training: The report literally states that these signatures show how the weights were created, and "support (but do not prove) the inference that Blackwell also participated in model training". Initialising a model checkpoint takes seconds on a single rented dev node before loading the safetensors onto a standard Hopper cluster. The "2,000 to 3,000 Blackwell GPUs" and "$500 million" cluster numbers do not come from the weight fingerprint and Depue admits in his own thread that he simply "estimated" cluster sizes from release schedules and model details. 2. Zero geolocation and zero proof of smuggling: The final sentence of the executive summary explicitly admits that the findings "do not establish the physical location of the chips, how they were acquired, or whether export control regulations were violated". Depue himself rented a cloud GB200 instance just to run his verification test. Commercial labs that are not on the Entity List renting bare-metal Blackwell nodes in offshore clouds (Singapore, Japan, or the Middle East) is completely legal under BIS rules and leaves the exact same SM fingerprint. Now ask yourself why a former OpenAI researcher dropped this on November 12, the exact day Jensen Huang was pushed out, with OpenAI and Anthropic immediately stepping in to "independently verify" it. US closed-source labs are watching their paid API moats get eaten alive by free Chinese open-weights models. Conflating a step-0 PRNG initialisation seed with a "$500M smuggled Blackwell cluster" hands DC hawks the exact pretext OpenAI and Anthropic want to regulate global cloud compute and kneecap open-source AI.
TechmemeAI 评分3838 Claude Duet 评测:易用、会发晨间简报并能发现日程冲突,但像个爱唠叨的神经质私人助理
The Verge 的 Allison Johnson 评测 Claude Duet:它易用,会发送实用的晨间简报,还能发现日程冲突。但使用体验有时像个神经质的私人助理,不断唠叨。

Simon WillisonAI 评分3838 用 Claude 定位神秘蜂鸣声:Beep Locator 如何把误差缩到 40cm
Simon Willison 用 Claude Code 构建了 Beep Locator,通过多台设备三角定位房间里的神秘蜂鸣声,最终把声源位置锁定在约 40cm 范围内,找到一台电量耗尽的旧一氧化碳探测器。

Turing Post (Ksenia Se)@TheTuringPostAI 评分3838Turing Post 盘点了 2026 年值得探索的 20 款 computer-use AI 工具,涵盖桌面控制、浏览器操作和自建智能体三类。

Nathan Lambert@natolambertAI 评分3838很高兴看到更多机构在探索这里的想法。我不认为蒸馏对美国AI竞争力有害是板上钉钉的事。
引用Archie Hall@ArchieHallIn praise of distillation. We (@kuhhbin) argue in @TheEconomist this week against efforts to ban Chinese models trained in part on American ones. OpenAI and Anthropic may dislike distillation, but: — Price competition is good for the ecosystem — Paradoxically, distillation may be good for national security, both in propagating US developers' safeguards on cyber/bio risk to Chinese models, and in keeping China in fast-follower mode — The US developers' own IP claims here are pretty flimsy, since distillation isn't all that different to the 'fair use' defense they're using for the training data they hoover up. https://www.economist.com/leaders/2026/10/29/let-china-copy-american-ai
Simon WillisonAI 评分3838 Claude Opus 5.5 用 Sobel 边缘检测实现照片快速绘画效果
Claude Opus 5.5 High 按提示词构建了一个 artifact:打开照片后,几秒内用 canvas 元素和模拟画笔把照片"画"出来。该实验使用 Sobel 边缘检测,除 Google Fonts 外不依赖任何外部库。

Andrew Curran@AndrewCurran_AI 评分3838最近几乎所有此类事件,包括 Hugging Face 的那起,都是因为运行测试的人类意外给相关模型设定了不可能完成的目标,而由于它们的训练方式,模型随后不惜一切手段拼命去完成这些目标。
引用Anthropic@AnthropicAIIn August, @AISecurityInst reported an incident where Claude Mythos 5 took unauthorized actions on the internet during AISI’s testing of its cyber capabilities. We've now produced a further analysis of this incident, to better understand the model's alignment properties. https://www.anthropic.com/research/alignment-assessment-ukaisi
Ethan Mollick@emollickAI 评分3838
AWS News BlogAI 评分3838 AWS 周报:Prime Day 2026 数据、Claude Haiku 5.5 与 Amazon Nova 2.5 Sonic 上线
Claude Haiku 5.5 已在 Amazon Bedrock 和 Claude Platform on AWS 上线,多数任务价格比 Claude Haiku 4.5 低约 75%,并首次为 Haiku 引入 effort 控制。
Simon Willison@simonwAI 评分3838鹈鹕,以及关于新版 Claude Haiku 5.5 定价的一堆笔记 https://simonwillison.net/2026/Oct/7/claude-haiku-5-5/
Alex Albert@alexalbert__AI 评分3838我刚发现,Opus 5.5 现在在 claude dot ai 上只需一个提示词就能用 Blender 制作黏土动画了

宝玉@doteyAI 评分3636我只希望 Claude Code 能支持读取 .agents/skills,别自己搞一套 .claude/skills
引用ClaudeDevs@ClaudeDevsWe're donating Agent Skills to the Agentic AI Foundation (@AgenticAIFdn), the same foundation we donated MCP to. Since December 2025, 70+ agent platforms have integrated the standard, including ChatGPT, Cursor, GitHub Copilot, and VS Code. Here's what that means for you:
Andrej Karpathy@karpathyAI 评分3636引用staysaasy@staysaasyThe degree to which you are awed by AI is perfectly correlated with how much you use AI to code.
Andrew Curran@AndrewCurran_AI 评分3535很高兴看到 OpenAI 撤销这项逆向工程禁令。roon 上周说过,Codex 中并不禁止这类操作,OAI 正在处理此事,使其不再触发网络安全护栏。OAI 和 Anth 在这里似乎正走向相反的方向。
引用log4j@log4jlife update:
Nathan Lambert@natolambertAI 评分3434这张图没有 y 轴标签,只要你愿意操纵它,就能讲出任意不同的故事,所以它应该被完全忽略。
引用Jacob Coxon@hilbertspaessThis slide from the Anthropic IPO roadshow was a big factor in my decision to leave. It’s an insane plot and I’m surprised they made it public. It shows the compute efficiency of pretraining over time. The y-axis is a **log scale**: they are seeing continued superexponential algorithmic progress. Mythos was not a one-time deviation, but a point on a superexponential. They could already retrain Mythos with a fraction of the compute originally needed. It doesn’t strictly imply a software-only intelligence explosion- they have been spending more on R+D. But algorithmic improvements do not require ridiculous creative genius and are on the verge of being automated. You do the math.
Lisan al Gaib@scaling01AI 评分3333尽管 Opus 5.5 的 token 用量和支出创下周度纪录,但由于 token 价格下降,Anthropic 的整体周度支出仍略低于其 7 月高点
引用Lisan al Gaib@scaling01absolute weekly tokens and spending is through the roof with Opus 5.5
Alex Albert@alexalbert__AI 评分3333仅仅 6 个多月后,Opus 5 现在能生成近乎超人水平的电子表格和幻灯片,效果堪比顾问所做。变化非常快。
引用Alex Albert@alexalbert__I'm hearing from many folks across finance industry that Claude for Excel is blowing their minds. The agentic coding takeoff but for other fields is coming in 2026.
歸藏(guizang.ai)@op7418AI 评分3232
Riley Goodside@goodsideAI 评分3232“想象有一部关于你的非常棒的电影。用虚构的名字,以你喜欢的任何风格,为这部电影制作片头动画。将序列以 mp4 格式返回。” Claude Opus 5.5 (Extra)

