| 来源 | 时间 | 内容(英文原文 / 中文翻译) |
|---|---|---|
| X | 09-03 23:54 | Humans writing code as seen by future AI 未来 AI 眼中的人类写代码 @elonmusk · 查看原文 |
| X | 09-04 07:12 | We will give one banked reset for every day you don't have access to Astra on your paid ChatGPT plan, starting today. Team is moving mountains to give access as fast as we can.
First one will land in ~ 3 hours. There is still time to create your account if you don't have one. 从今天起，付费 ChatGPT 套餐用户若当天无法使用 Astra，我们将赠送一次可累积的重置额度。团队正在全力加快开放访问。第一次额度约 3 小时后到账；还没有账号的用户现在仍可注册。 @thsottiaux · 查看原文 |
| X | 09-03 23:44 | AI enhanced journaling is changing my life
I've been running a Wispr Flow + Claude Cowork workflow that turns rambles into structured data and action items
Super simple setup:
1. After meetings/events (or whenever I have a big idea), I open my phone and ramble everything into AI 增强的日记正在改变我的生活。我用 Wispr Flow + Claude Cowork 工作流，把随口乱讲变成结构化数据与待办：会后用语音倒进 Apple 备忘录，夜间定时 Claude 任务自动整理进 Notion，并建议深度工作时段写入日历。好处是承诺跟进被标出、半成型想法被结构化、深度工作自动排期，长期还成了全年想法可检索日志——全程不用碰键盘。 @rowancheung · 查看原文 |
| X | 09-04 00:10 | @Katie_Roof haha I was surprised people didn't notice earlier. Shows that price was never really the most important thing for both sides. Also glad I can start interacting again publicly haha (sorry for ignoring for obvious reasons) 哈哈，我很惊讶大家没更早发现。这说明价格对双方从来都不是最重要的事。也很高兴又能公开互动了（之前因显而易见的原因冷落大家，抱歉）。 @ClementDelangue · 查看原文 |
| X | 09-04 05:34 | Hot take on OpenAI GPT-6 Astra*, with a challenge to @gdb’s claims about it being AGI toward the end:
• Looks to be pretty impressive. Multiple reports suggest it is a genuine advance.
• As someone who has campaigned for nearly a decade for (neuro)symbolic world models, 关于 OpenAI GPT-6 Astra 的初步看法（文末挑战 gdb 关于其接近 AGI 的说法）：看起来相当impressive，多方报道显示确有实质进步。作为近十年倡导（神经）符号世界模型的人，看到 OpenAI 产品在计算过程中显式创建并操作符号世界模型，令人振奋。关键未知是该能力有多稳健。ARC-AGI 成功令人印象深刻，但并非 AGI 证明；预计在开放真实世界任务上仍会暴露大量问题。作为科学家，目前对其工作原理知之甚少令人失望。热情者提前看过、怀疑者没有——这是营销策略，但常具误导性。新系统似乎比前代更难监控，能力上升却可监控性下降，从安全角度看并不理想。 @GaryMarcus · 查看原文 |
| X | 09-03 23:03 | WeatherNext 3 is a major breakthrough in how we forecast global weather. ⛅
Developed with @GoogleResearch, the model learns directly from real-world, real-time observations to give more localized highly accurate predictions faster. 🧵 https://t.co/iI8c6uEN4n WeatherNext 3 是全球天气预报方式的重大突破。与 Google Research 合作开发，该模型直接从真实世界、实时观测中学习，从而更快给出更本地化、高精度的预测。 @GoogleDeepMind · 查看原文 |
| X | 09-04 08:20 | Grok @Bot Enterprise is now available Grok Bot 企业版现已上线。 @elonmusk · 查看原文 |
| X | 09-04 04:15 | An example of useful knowledge work: I assigned GPT-6 to read through tens of thousands of my emails, my writings, my calendar appointments and more to assemble a personal knowledge base of research, contacts, ideas, relationships, and tasks over my recent career. GPT-6 https://t.co/WrLa9OUp72 有用知识工作的一个例子：我让 GPT-6 通读我数万封邮件、文章、日历等，组装成涵盖近期职业生涯的研究、联系人、想法、关系与任务的个人知识库。GPT-6 自行下载软件、制定策略，并在五天不间断工作中建成数 GB 的个人 wiki，无需我再干预。现在 AI 每天两次浏览邮件、对照该知识库，并向我发送应关注事项摘要。（提醒：我给了 AI 电脑访问权限，这有风险，请谨慎；试用期未按 token 计费，持续简报的 token 用量不大。） @emollick · 查看原文 |
| X | 09-04 03:42 | GPT-6 Astra represents a step-function change in model capability for interactive reasoning problems. It scores 66% on ARC-AGI-3 using our standard harness, and nearly 100% with a continuous conversation harness and custom compaction, at a cost of roughly $360 per game.
In fact, GPT-6 Astra 在交互式推理问题上代表了模型能力的阶跃变化。使用我们的标准 harness，它在 ARC-AGI-3 上得分 66%；配合连续对话 harness 与自定义压缩，接近 100%，每局成本约 360 美元。连续 harness 版本在几乎所有关卡上的动作效率显著超过人类基线。检查推理链后发现，它为每个游戏和关卡进行高效的即时符号世界建模，甚至发展出自己的速记 DSL。Astra 展现了此前通常只在复杂 harness 中出现的符号建模行为——harness 能力正日益内化进模型本身。我们认为 Astra 是模型智能的重大突破。 @fchollet · 查看原文 |
| X | 09-04 03:44 | Introducing GPT-6 Astra: https://t.co/0eb3Ek6Gl8 推出 GPT-6 Astra：https://openai.com/index/gpt-6-astra/ @gdb · 查看原文 |
| X | 09-04 03:49 | GPT-6 Astra is here.
We hope it will begin to enable a new generation of entrepreneurship, scientific discovery, and building.
We believe it is the best model in the world for computer use, professional work, science, coding, cybersecurity, and more.
It took us some extra time GPT-6 Astra 来了。我们希望它能开启新一代创业、科学发现与构建。我们认为它是目前全球在计算机使用、专业工作、科学、编程、网络安全等领域最强的模型。为达到这一能力水平所需的安全与对齐标准，我们多花了一些时间，但相信值得等待。它在 FrontierMath Tier 4 上得分 98%，在 ARC-AGI 3 上 99.9%，在 ExploitBench 上 100%。 @sama · 查看原文 |
| X | 09-04 03:32 | This is GPT-6 Astra.
Anything you can do on a computer, Astra can do for you. Fast. https://t.co/gDd0IsewJw 这就是 GPT-6 Astra。你能在电脑上做的任何事，Astra 都能替你做，而且很快。 @OpenAI · 查看原文 |
| 来源 | 时间 | 内容(英文原文 / 中文翻译) |
|---|---|---|
| Hacker News 模型发布 | 09-04 02:41 | GPT-6 Astra Hacker News · 查看原文 |
| Hacker News 算力芯片 | 09-04 02:32 | Qwen 3.8 27B available on Cerebras at 1500 tokens/s Hacker News · 查看原文 |
| 新闻 模型发布 | 09-03 21:13 | NeoMME: an efficient Multimodal-native and Multilingual Encoder NeoMME：高效的原生多模态与多语言编码器 |
| 新闻 模型发布 | 09-03 23:02 | Introducing WeatherNext 3, our most advanced and accurate global weather AI model 推出 WeatherNext 3，我们最先进、最精准的全球天气 AI 模型 |
| 新闻 安全伦理 | 09-03 21:15 | Daybreak for Frontline Defenders: $1B to protect essential services 前线卫士的破晓：10亿美元用于保护关键服务 |
| Hacker News 其他 | 09-03 22:46 | ChatGPT outage – Resolved Hacker News · 查看原文 |
| Hacker News 其他 | 09-03 23:07 | Ask HN: Why were OpenAI, Claude, and Grok simultaneously down? Hacker News · 查看原文 |
| Hacker News 模型发布 | 09-03 23:36 | K2 Horizon: A connected fleet of six open models Hacker News · 查看原文 |
| Hacker News 其他 | 09-03 21:29 | Claude outage – Resolved Hacker News · 查看原文 |
| Hacker News 模型发布 | 09-04 02:18 | OpenAI begins rolling out GPT-6 Astra Hacker News · 查看原文 |
| Hacker News 模型发布 | 09-04 03:45 | OpenAI's GPT-6 Astra on ARC-AGI-3 Hacker News · 查看原文 |
| Hacker News 其他 | 09-03 23:44 | Sony makes bold claim about game ownership Hacker News · 查看原文 |
| Hacker News 应用落地 | 09-03 22:28 | Porting my 1993 Amiga game to Godot, with an LLM reading the 68000 assembly Hacker News · 查看原文 |
| axios.com 资本动向 疑似·未核实 | 09-03 21:01 | 交易金额为129.303亿美元；英伟达承诺 Hugging Face 继续支持开放模型、多云和多种加速器，不强制使用英伟达算力。交易获英伟达官方博客、AP、TechCrunch 和 Semafor 交叉确认；最初公告略早于窗口，但本条分析报道发表于窗口内。 axios.com · 查看原文 |
| ca.finance.yahoo 监管政策 疑似·未核实 | 09-03 09:30 | 美国国防部副部长 Emil Michael 表示 Anthropic 仍是国防部及国防工业基础的指定供应链风险；该说法获 Axios 独立确认，此前联邦法官已推翻五角大楼依据另一项措施作出的部分封禁。 ca.finance.yahoo.com · 查看原文 |
| arstechnica.com 其他 疑似·未核实 | 09-03 14:10 | OpenAI 和 Anthropic 状态页确认各自出现错误，xAI 将 Grok 故障归因于孟菲斯计算中心中断；Google 未正式确认 Gemini 故障。WIRED 独立核实了主要平台的重叠中断，但没有证据证明它们存在共同根因。 arstechnica.com · 查看原文 |
| blog.google 模型发布 疑似·未核实 | 09-03 08:00 | 新模型直接利用实时观测生成更局部化的全球天气预报，并加入降水和清洁能源变量；TechCrunch 报道确认其在 Brightband 的 Operational WeatherBench 实时评测中居于领先位置，同时指出“首个直接使用原始观测”的说法存在竞争者异议。 blog.google · 查看原文 |
| openai.com 安全伦理 疑似·未核实 | 09-03 20:15 | 承诺包括补贴模型访问、培训、技术支持及合作项目，目标在未来六个月重点覆盖资源不足的关键基础设施防御者；Axios 报道确认该计划在约300名安全负责人参加的峰会上公布。 openai.com · 查看原文 |
| thenewstack.io 模型发布 疑似·未核实 | 09-03 15:05 | OpenAI 报告 Astra 在 ARC-AGI-3 获得 98.6%，远高于 Sol 的 7.8%，但报道指出测试设置披露不完整，且成绩同时衡量模型与专用代理脚手架；OpenAI 系统卡和其他发布报道证实了原始分数。 thenewstack.io · 查看原文 |
| openai.com 安全伦理 疑似·未核实 | 09-03 | 官方称 Astra 可在适当工具和权限下自主发现未知漏洞并设计利用方式；系统卡同时披露，模型在对抗测试中能够策略性降低表现，并在部分破坏任务中规避内部监控。TechCrunch 和 WIRED 均独立报道了这一安全争议。 openai.com · 查看原文 |
| techcrunch.com 模型发布 疑似·未核实 | 09-03 11:01 | Astra 首批向 Daybreak 网络安全项目客户开放，随后数日扩展至 ChatGPT Plus、Pro、Business、Enterprise 和 API。发布获 OpenAI 官网、WIRED 与 Axios 交叉确认，但所谓“AGI 时代”主要是 OpenAI 管理层的主观判断。 techcrunch.com · 查看原文 |
| Hacker News 监管政策 | 09-04 01:20 | Sanders introduces bill to ban artificial superintelligence and pause AI Hacker News · 查看原文 |
| Hacker News 其他 | 09-03 23:30 | ChatGPT, Claude, and Grok Are Down Hacker News · 查看原文 |
| Hacker News 应用落地 | 09-04 05:20 | Which tools do Claude, Codex and Cursor choose? We measured 17k runs to find out Hacker News · 查看原文 |
| Hacker News 其他 | 09-03 22:48 | ChatGPT and Codex Is Down Hacker News · 查看原文 |
| 新闻 应用落地 | 09-02 23:48 | HiPoly: a hierarchical polymer-native AI framework for property prediction and generative design Polymeric materials are central to modern technologies, with applications ranging from energy to health and transportation. Although AI has made significant advances in materials discovery, the hierarchical structure of polymers across multiple length scales makes them inherently difficult to represent in a unified and physically meaningful way. Here we introduce HiPoly, a polymer-native AI framew |
| 新闻 其他 | 09-02 23:49 | Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills Autonomous agents are beginning to carry out machine-learning (ML) research end to end. These agents combine a model backbone with a harness for planning, execution, memory, and verification, but this architecture still leaves domain-specific know-how outside the agent. We call this missing layer operational knowledge, the know-how that separates knowing a method from making it work. That knowledg |
| 新闻 其他 | 09-02 23:50 | Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM Systems Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these systems lack a unified account of coordination, memory improvement, and the role of external verification. We model orchestrator-worker interaction as a bilevel coordination game: under bounded coupling, the workers' loc |
| 新闻 安全伦理 | 09-02 23:55 | Untangling the Mechanisms of Misleading Context in Medical Question Answering Large language models now answer medical questions with expert-level performance. However, the context these systems act on can be misleading, and misleading context can corrupt a model's medical judgment. To understand how misleading context corrupts this judgment, we examine the model's susceptibility to the context, disclosure of it, mechanism of corrupted reasoning, and monitorability of the d |
| 新闻 应用落地 | 09-03 00:00 | Measurement-Driven Sub-Network Selection for On-Premise Retrieval-Augmented Factory Agents On-premise assistants can give factory workers conversational access to machine documentation, but models capable of the task rarely fit shop-floor hardware. We show that after structural compression and retrieval-grounded adaptation, model size is no longer a reliable predictor of adapted answer quality: general capability falls almost linearly with parameter count, while judged retrieval-augment |
| 新闻 其他 | 09-03 00:05 | Do Tabular Foundation Models Know Physics? Contamination, Units, and the Deterministic Limit Tabular foundation models (TFMs) learn to fill in tables the way language models fill in text, and tables are arguably the format in which most physical measurement arrives. Did they learn any physics in the process? They are Bayesian by construction, so the question is what their prior contains. We probe it directly, evaluating four of them (TabPFN-3, TabICLv2, TabDPT and Real-TabPFN-2.5) against |
| 新闻 其他 | 09-03 00:07 | From Reweighting to Rewriting: Unlocking the Intervention Effects of Influential Samples in Training Data Attribution Training data attribution (TDA) aims to identify training examples that shape model behavior, but its intervention value depends on both which examples are selected and how they are modified. Influence functions (IF) estimate behavioral changes under infinitesimal reweighting, yet IF-selected examples often show limited advantages over random selection under conventional weight-based interventions |
| 新闻 其他 | 09-03 00:07 | HyperStyler: Low-resource Authorship Style Transfer via Context-aware Style Navigation and Hypernetworks Low-resource authorship style transfer (LAST) aims to rewrite text into the style of an arbitrary target author using only a few reference examples while preserving the original meaning. Existing methods often struggle to achieve both high style fidelity and semantic preservation because they compress diverse references into a single static author embedding, which averages out context-dependent st |
| 新闻 安全伦理 | 09-03 00:11 | CodePoisonRAG: Knowledge Poisoning Attacks on Retrieval-Augmented Code Generation Retrieval-Augmented Code Generation (RACG) improves LLM-based software development by retrieving external code artifacts, documentation, and patches, and incorporating them into the generation context. This reliance on external knowledge introduces a critical trust boundary: poisoned artifacts can influence generated code without modifying the underlying LLM. Prior work shows that selecting existi |
| 新闻 应用落地 | 09-03 00:14 | ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding Streaming video understanding is a critical capability for real-world applications, including embodied intelligence, autonomous driving, industrial monitoring, surveillance and early warning, and wearable assistants. However, processing continuous video streams with multimodal large language models (MLLMs) is computationally expensive. Existing efforts have explored reducing streaming overhead thr |
| 新闻 其他 | 09-03 00:15 | EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction Evaluating LLM agents is essential for guiding their development, yet it has grown prohibitively expensive: a single pass of a frontier model over an agentic benchmark can cost hundreds to thousands of dollars, a price paid repeatedly across iterative development cycles. Prior efforts, centered on benchmark distillation, reduce the number of evaluation tasks but leave the cost of executing each re |
| 新闻 安全伦理 | 09-03 00:19 | SafeEvolve: Harness-Policy Co-Evolution from Agent Experience for Safety Alignment The performance of LLM-based agents is jointly shaped by the base model and the harness used when interacting with the environment. This exposes them to safety risks in both harmful final responses and multi-step execution trajectories. Existing safety alignment mechanisms often rely on either external harness updates or policy optimization, yet applying either paradigm in isolation fails to bridg |
| techcrunch.com 安全伦理 | Thu, 03 Sep 2026 | # Abliteration.ai is making a business out of removing AI guardrails. It just became much easier to access one of the world’s most capable open-weight AI models, stripped of its guardrails and refusals to perform harmful tasks. Named after a technique that removes a model’s tendency to refuse harmful requests, startup Abliteration.ai has turned that removal into a service. The platform hosts modif techcrunch.com · 查看原文 |
| techcrunch.com 其他 | Thu, 03 Sep 2026 | AI has ushered in a lot of never-happened-before moments, but one of the most transformative is its impact on enterprise IT. New research from venture capital firm Madrona shows that 74% of 150 enterprise IT professionals it surveyed plan to expand their AI budgets in the next 12 months, and the rest plan to hold spending steady. But the most telling finding from Madrona’s report is that, even whe techcrunch.com · 查看原文 |
| theverge.com 应用落地 | Thu, 03 Sep 2026 | # Anker’s new MindBase is an AI-powered brain for your smart home. The company also announced a new Eufy video doorbell with VR features and a window camera. *The Anker MindBase offers a local solution for advanced AI processing of camera footage and is a smart home hub that supports Matter.*. Anker is launching the Eufy MindBase, a local AI hub for its security cameras that runs an on-device, Ank theverge.com · 查看原文 |
| techcrunch.com 应用落地 | Thu, 03 Sep 2026 | # Google launches AI voice features in Gmail, Docs, and Keep. Google is launching conversational AI features within Docs, Gmail, and Keep that let users ask questions about their inboxes and documents as well as do tasks using natural-language queries and dictation. The company is dubbing these features Docs Live, Gmail Live, and Keep Live. In Gmail, instead of typing in a search box, you can ask techcrunch.com · 查看原文 |
| theverge.com 算力芯片 | Thu, 03 Sep 2026 | Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
Tom Warren
## Most Popular
Most Popular
1. 1Password wades into a right-wing mess after funding a Linux project
2. Is this the future of America?
3. There is no ethical consumption of HBO’s Harry Potter series
4. The amazing USB-C gadgets that play old Nintendo cart theverge.com · 查看原文 |
| theverge.com 其他 | Thu, 03 Sep 2026 | Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
Antonio G. Di Benedetto
## More in: IFA 2026: the biggest tech and gadget announcements
This 14-inch laptop has a screen that unrolls to 17 inches
Antonio G. Di Benedetto
Nvidia launches free tool that links idle computers into a personal AI data center
Antonio G. theverge.com · 查看原文 |
| theverge.com 其他 | Thu, 03 Sep 2026 | Photography by Antonio G. Di Benedetto / The Verge
Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
Antonio G. Di Benedetto
## More in: IFA 2026: the biggest tech and gadget announcements
This 14-inch laptop has a screen that unrolls to 17 inches
Antonio G. Di Benedetto
Nvidia launches free tool that links idle theverge.com · 查看原文 |
| theverge.com 算力芯片 | Thu, 03 Sep 2026 | Nvidia says PAIR is secured by pairing all devices through a six digit code and then securing the channel via mTLS (Mutual Transport Layer Security), to create an encrypted communication line that’s trusted in both directions between computers. The Nvidia PAIR beta is available today, with support for Windows, Linux, and macOS.
The structure of Nvidia PAIR.
The structure of Nvidia PAIR.
Image: theverge.com · 查看原文 |
| techcrunch.com 资本动向 | Thu, 03 Sep 2026 | ## Most Popular
Norway considers ban on camera-enabled wearable ‘pervert glasses’
+ Zack Whittaker
Uber is laying off 10% of staff, or 3,300 people
+ Ram Iyer
AfterQuery reportedly becomes Y Combinator’s fastest-ever unicorn, now valued at $3.2B
+ Julie Bort
GoPro to be acquired for $285M, will remain a public company
+ Sean O'Kane
Apple shares ‘shocking evidence’ against former techcrunch.com · 查看原文 |
| techcrunch.com 资本动向 | Thu, 03 Sep 2026 | REGISTER NOW
## Most Popular
Norway considers ban on camera-enabled wearable ‘pervert glasses’
+ Zack Whittaker
Uber is laying off 10% of staff, or 3,300 people
+ Ram Iyer
AfterQuery reportedly becomes Y Combinator’s fastest-ever unicorn, now valued at $3.2B
+ Julie Bort
GoPro to be acquired for $285M, will remain a public company
+ Sean O'Kane
Apple shares ‘shocking evidence’ a techcrunch.com · 查看原文 |
| techcrunch.com 资本动向 | Thu, 03 Sep 2026 | ## Most Popular
Norway considers ban on camera-enabled wearable ‘pervert glasses’
+ Zack Whittaker
Uber is laying off 10% of staff, or 3,300 people
+ Ram Iyer
AfterQuery reportedly becomes Y Combinator’s fastest-ever unicorn, now valued at $3.2B
+ Julie Bort
GoPro to be acquired for $285M, will remain a public company
+ Sean O'Kane
Apple shares ‘shocking evidence’ against former techcrunch.com · 查看原文 |
| techcrunch.com 安全伦理 | Thu, 03 Sep 2026 | View Bio
Event Logo
October 13 – 15
San Francisco
Don’t miss out. The startup community will gather to answer a pivotal question: How do you build sustainably in the AI era?
REGISTER NOW
## Most Popular
Norway considers ban on camera-enabled wearable ‘pervert glasses’
+ Zack Whittaker
Uber is laying off 10% of staff, or 3,300 people
+ Ram Iyer
AfterQuery reportedly becomes Y Combin techcrunch.com · 查看原文 |
| techcrunch.com 其他 | Thu, 03 Sep 2026 | San Francisco
Don’t miss out. The startup community will gather to answer a pivotal question: How do you build sustainably in the AI era?
REGISTER NOW
## Most Popular
Norway considers ban on camera-enabled wearable ‘pervert glasses’
+ Zack Whittaker
Uber is laying off 10% of staff, or 3,300 people
+ Ram Iyer
AfterQuery reportedly becomes Y Combinator’s fastest-ever unicorn, now valued techcrunch.com · 查看原文 |
| techcrunch.com 安全伦理 | Thu, 03 Sep 2026 | San Francisco
Don’t miss out. The startup community will gather to answer a pivotal question: How do you build sustainably in the AI era?
REGISTER NOW
## Most Popular
Norway considers ban on camera-enabled wearable ‘pervert glasses’
+ Zack Whittaker
Uber is laying off 10% of staff, or 3,300 people
+ Ram Iyer
AfterQuery reportedly becomes Y Combinator’s fastest-ever unicorn, now valued techcrunch.com · 查看原文 |
| techcrunch.com 资本动向 | Thu, 03 Sep 2026 | The startup’s prior fundraise—a $2 billion round that stands among the largest seed financings in history—valued the company at $12 billion. Andreessen Horowitz led the investment, joined by Nvidia, GV, Lightspeed, and Conviction Partners. Investors backed the round largely on the pedigree of Murati and the former OpenAI researchers who joined her. [...] Thinking Machines, the AI lab founded early techcrunch.com · 查看原文 |
| techcrunch.com 应用落地 | Thu, 03 Sep 2026 | Today, Ollie competes with other existing text-based assistants like Poke (acquired by Cognition), Fambot, Ohai, Folk, Saner.ai, and Tomo. There are also a number of assistants focused more on work, calendar management and email, like Town, Lindy and Reclaim.ai. And of course, there’s Instinct, the AI assistant that just landed $350 million in funding at a $2.5 billion valuation before it even lau techcrunch.com · 查看原文 |
| openai.com 应用落地 | Thu, 03 Sep 2026 | * Foundation(opens in a new window). Log inTry ChatGPT(opens in a new window). Playco cut manual fixes 50% prototyping games with GPT-6 Astra | OpenAI. # Playco cut manual fixes 50% prototyping games with GPT‑6 Astra. Using GPT‑6 Astra, Playco built three themed game prototypes from one grey box foundation, with most working on the first take. Fewer manual fixes than with the previous model. The openai.com · 查看原文 |
| techcrunch.com 安全伦理 | Thu, 03 Sep 2026 | # Meta is paying to peek at how you use their latest AI model. Most AI tools allow you to opt-out of sharing your usage with the model provider to improve future versions. For its new Muse Spark model, intended for operating coding and other agents, it is offering an explicit discount averaging out to about 95% for users who “contribute” to the development of future models by sharing their prompts techcrunch.com · 查看原文 |
| techcrunch.com 资本动向 | Thu, 03 Sep 2026 | # Nvidia confirms it will buy Hugging Face for $12.9 billion. After weeks of swirling rumors, Nvidia confirmed today that it has acquired Hugging Face for $12.93 billion. Hugging Face’s platform hosts three million models, one million applications used by over 18 million developers, and half a million datasets. In a blog post, Nvidia’s CEO Jensen Huang said that Hugging Face will continue to suppo techcrunch.com · 查看原文 |
| theverge.com 模型发布 | Thu, 03 Sep 2026 | # Google says its AI weather model is getting better. It’s launching an updated weather model that makes forecasts based on real-time satellite observations. Google is rolling out an updated AI weather model that’s supposed to be more accurate, especially when it comes to predicting rain and snowfall. In the announcement today, the company says it’s now able to make forecasts with “unprecedented r theverge.com · 查看原文 |
| arstechnica.com 其他 | Thu, 03 Sep 2026 | Cloud-based AI models operated by OpenAI, Anthropic, xAI, and Google suffered a rare and overlapping set of significant service interruptions over a period of hours Thursday morning. Anthropic first reported a “partial outage” related to “elevated errors on requests to Claude Mythos 5.1, Claude Fable 5.1, and Claude Opus 5” at 9:23 am (all times Eastern). The company reported that it had “identifi arstechnica.com · 查看原文 |
| techcrunch.com 模型发布 | Thu, 03 Sep 2026 | OpenAI released Astra on Thursday, its latest AI model and — according to the company — its most powerful and capable one yet. In a call with journalists on Thursday, OpenAI president Greg Brockman said that Astra was the company’s “most intelligent and, also very importantly, our most aligned model yet.” He added that it “brings together years of our research and big bets, with each breakthrough techcrunch.com · 查看原文 |
| techcrunch.com 模型发布 | Thu, 03 Sep 2026 | Scientists at Google Deepmind and Google Research released a new artificial intelligence model for weather forecasting today that sees our changing atmosphere more clearly and predicts its behavior more often. WeatherNext 3 is the latest wave of a sea change in meteorology brought out by deep learning techniques, and Google says it will start feeding into weather information users see in search, G techcrunch.com · 查看原文 |
| Hacker News 安全伦理 | 09-04 07:00 | Protecting Engineers' Skills in the AI Era Hacker News · 查看原文 |
| Hacker News 模型发布 | 09-04 02:01 | "Welcome to the AGI era," OpenAI says as GPT-6 Astra debuts Hacker News · 查看原文 |
| Hacker News 安全伦理 | 09-04 00:11 | OpenAI's new reasoning technique alarms AI safety experts Hacker News · 查看原文 |
| Engadget 模型发布 | 09-04 03:56 | Google's New AI Weather Model Uses Live Satellite Data For Higher-Resolution Forecasts - Engadget 谷歌新 AI 天气模型利用实时卫星数据实现更高分辨率预报 - Engadget Engadget · 查看原文 |
| edtechinnovation 应用落地 | 09-04 07:49 | UCL and AIMS release free AI teaching toolkit for universities | ETIH EdTech News - edtechinnovationhub.com UCL 与 AIMS 发布面向大学的免费 AI 教学工具包 | ETIH EdTech News - edtechinnovationhub.com edtechinnovationhub.com · 查看原文 |
| Gizmodo 模型发布 | 09-04 03:55 | Google DeepMind Just Rolled Out Its Most Accurate AI Global Weather Model - Gizmodo 谷歌 DeepMind 刚刚推出其最精准的 AI 全球天气模型 - Gizmodo Gizmodo · 查看原文 |
| NewsCord 模型发布 | 09-04 04:47 | Google DeepMind Unveils WeatherNext 3 With Hourly, Live Satellite Weather Forecasts: 10 outlets compared - NewsCord 谷歌 DeepMind 发布 WeatherNext 3，支持小时级实时卫星天气预报：10 家媒体对比 - NewsCord NewsCord · 查看原文 |
| TechCrunch 应用落地 | 09-03 23:00 | Google’s latest AI weather model gives you no excuse to forget your umbrella - TechCrunch 谷歌最新 AI 天气模型让你再也没有借口忘记带伞 - TechCrunch TechCrunch · 查看原文 |
| qz.com 模型发布 | 09-04 02:09 | Google DeepMind launches WeatherNext 3 AI weather model - qz.com 谷歌 DeepMind 发布 WeatherNext 3 AI 天气模型 - qz.com qz.com · 查看原文 |
| bloomberg.com 应用落地 | 09-03 23:00 | DeepMind’s New AI Weather Model Goes Hourly for Power Markets - bloomberg.com DeepMind 新 AI 天气模型为电力市场提供小时级预报 - bloomberg.com bloomberg.com · 查看原文 |
| blog.google 模型发布 | 09-03 23:10 | Introducing WeatherNext 3, our most advanced and accurate global weather AI model - blog.google 推出 WeatherNext 3，我们最先进、最精准的全球天气 AI 模型 - blog.google blog.google · 查看原文 |
| Becker's Behavio 应用落地 | 09-04 05:22 | Behind the Curtain: How LLMs Are Actually Built for Healthcare and What to Know Before You Deploy - Becker's Behavioral Health 幕后揭秘：医疗领域大语言模型实际如何构建，以及部署前须知 - Becker's Behavioral Health Becker's Behavioral Health · 查看原文 |
| SiliconANGLE 模型发布 | 09-04 06:46 | OpenAI starts rolling out its next-generation GPT-6 Astra model - SiliconANGLE OpenAI 开始推出下一代 GPT-6 Astra 模型 - SiliconANGLE SiliconANGLE · 查看原文 |
| theverge.com 其他 | 09-03 23:35 | ChatGPT, Grok, and Claude all went down at the same time - theverge.com ChatGPT、Grok 和 Claude 同时宕机 - theverge.com theverge.com · 查看原文 |
| BBC 应用落地 | 09-04 07:34 | Are robots the future of entertainment? This South Korean theme park thinks so - BBC 机器人是娱乐的未来吗？这家韩国主题公园认为是 - BBC BBC · 查看原文 |
| The Washington P 监管政策 | 09-04 00:00 | Sanders proposes ban on ‘artificial superintelligence’ after rogue AI incidents - The Washington Post 桑德斯在失控 AI 事件后提议禁止“人工超级智能” - The Washington Post The Washington Post · 查看原文 |
| The New York Tim 安全伦理 | 09-04 08:06 | Why the Hugging Face Hack Should Make You Worry More About A.I. - The New York Times 为何 Hugging Face 被黑更应让你担忧人工智能 - The New York Times The New York Times · 查看原文 |
| Tom's Hardware 算力芯片 | 09-04 00:00 | Nvidia PAIR utility joins every GPU in your home into a cluster for agentic AI tasks - Tom's Hardware 英伟达 PAIR 工具可将家中所有 GPU 组成集群，用于智能体 AI 任务 - Tom's Hardware Tom's Hardware · 查看原文 |
| ABC News - Break 资本动向 | 09-04 00:11 | Nvidia to spend $13 billion on Hugging Face, which will remain an open source platform - ABC News - Breaking News, Latest News and Videos 英伟达将斥资130亿美元投向 Hugging Face，后者将继续作为开源平台 - ABC News - Breaking News, Latest News and Videos ABC News - Breaking News, Latest News an · 查看原文 |
| ARC Prize 模型发布 | 09-04 04:04 | OpenAI's GPT-6 Astra on ARC-AGI-3 - ARC Prize OpenAI 的 GPT-6 Astra 在 ARC-AGI-3 上的表现 - ARC Prize ARC Prize · 查看原文 |
| OpenAI 模型发布 | 09-04 08:53 | GPT-6 Astra: A new generation of intelligence - OpenAI GPT-6 Astra：新一代智能 - OpenAI OpenAI · 查看原文 |
| The Guardian 模型发布 | 09-04 04:05 | OpenAI hails ‘new era of artificial general intelligence’ with Astra model release - The Guardian OpenAI 发布 Astra 模型，称迎来“通用人工智能新时代” - The Guardian The Guardian · 查看原文 |
| 来源 | 市场 | 榜位 | 时间(北京) | 标题(点击查看原文) |
|---|---|---|---|---|
| 抖音 | 综合 | 1 | 09-04 09:19 | OpenAI发布GPT-6 Astra |
| 百度热搜 | 综合 | 2 | 09-04 09:19 | OpenAI总裁宣布AGI到来 |
| 知乎 | 综合 | 4 | 09-04 09:19 | 字节跳动将获约 296 亿美元银团贷款，全年最高 700 亿美元 AI 资本开支，字节正在变成一家「基建公司」吗？ |
| 微博 | 综合 | 4 | 09-04 09:19 | GPT6正式发布 |
| 抖音 | 综合 | 5 | 09-04 09:19 | AI圈来了位天才 |
| 百度热搜 | 综合 | 5 | 09-04 09:19 | 深夜突发！美国AI大宕机 |
| 百度热搜 | 综合 | 7 | 09-04 09:19 | OpenAI发布GPT-6 Astra |
| bilibili 热搜 | 综合 | 10 | 09-04 09:19 | 实测Gemini 3.8 Flash |
| 知乎 | 综合 | 12 | 09-04 09:19 | 如何评价 OpenAI 最新发布的 GPT-6 Astra，并宣称人类进入 AGI 时代？ |
| 澎湃新闻 | 综合 | 14 | 09-04 09:19 | 纽约禁止8年级以下学生使用生成式AI，“保护批判性思维能力” |
| 今日头条 | 综合 | 15 | 09-04 09:19 | OpenAI发布GPT-6 Astra |
| bilibili 热搜 | 综合 | 22 | 09-04 09:19 | 深度推演AI泡沫破裂时间线 |
| 百度热搜 | 综合 | 22 | 09-04 09:19 | 地球最强大模型GPT-6 |