Guozhen AIGlobal AI field notes and model intelligence

Realtime AI News

Track model releases, AI agents, open-source projects, infrastructure, policy, and product updates as they publish.

Read details: OpenAI Publishes JioStar Case Study on Running AI Discovery at 500M-User Scale
OpenAI / JioStar / Deployment

OpenAI Publishes JioStar Case Study on Running AI Discovery at 500M-User Scale

OpenAI has published an interview with JioStar chief architect Vijay Seshadri on building AI-powered discovery for 500 million users. The conversation dwells less on model capability than on evaluations, guardrails and the unglamorous work of putting AI into production.

Read details: Hugging Face ships token pooling-aware training for multi-vector retrieval models
Hugging Face / Embedding / Open Source

Hugging Face ships token pooling-aware training for multi-vector retrieval models

Hugging Face has published a pooling-aware training recipe for multi-vector embedding models, along with a Sentence Transformers loss wrapper, training scripts and two ModernBERT-based checkpoints. In its NanoBEIR tests, a model trained this way retained 97.8% of its unpooled retrieval score at a 4x reduction in document vectors, up from 94.6% for ordinary training.

Hugging Face 开源 Token 池化感知训练:文档向量压缩4倍仍保留97.8%检索质量
Read details: Gemini 4 Carbon连夜上桌!力压Fable 5.2花费仅一半,还挖出了PHP同级漏洞… yeeyi yeeyi
AI News / Gemini / Models

Gemini 4 Carbon连夜上桌!力压Fable 5.2花费仅一半,还挖出了PHP同级漏洞… yeeyi yeeyi

Gemini 4 Carbon连夜上桌!力压Fable 5.2花费仅一半,还挖出了PHP同级漏洞… yeeyi yeeyi. Gemini 4 Carbon连夜上桌!力压Fable 5.2花费仅一半,还挖出了PHP同级漏洞… yeeyi yeeyi

Gemini 4 Carbon连夜上桌!力压Fable 5.2花费仅一半,还挖出了PHP同级漏洞… yeeyi yeeyi
Read details: New York's AI Safety Law Takes Effect, Putting OpenAI and Anthropic Under New Oversight
Policy / OpenAI / Anthropic

New York's AI Safety Law Takes Effect, Putting OpenAI and Anthropic Under New Oversight

New York is moving to enforce its sweeping AI safety law, requiring major developers such as OpenAI and Anthropic to register, disclose safety practices, and report critical incidents. Companies that fail to comply could face fines under the new state framework.

纽约州 AI 安全法生效,OpenAI 与 Anthropic 进入州级合规流程
Read details: All seven Anthropic co-founders will each cash out about $250M in the IPO
Anthropic / IPO / Business

All seven Anthropic co-founders will each cash out about $250M in the IPO

Anthropic's new S-1 filing shows each of its seven co-founders selling 3,378,378 shares, worth just under $250 million apiece at the $74 midpoint of a $70 to $78 range. The group will keep roughly 22% of Class A stock and, through supervoting Class F shares, 51.1% of the vote.

Anthropic七位联合创始人IPO中各套现约2.5亿美元
Read details: OpenAI rolls out Business Hub, making ChatGPT a little more like Google
OpenAI / ChatGPT / Business

OpenAI rolls out Business Hub, making ChatGPT a little more like Google

OpenAI launched a beta Business Hub on November 10 where companies can build verified profiles, deploy a dedicated business agent, and see metrics such as profile views, mentions and website clicks. It marks a shift from in-app shopping toward a conversational discovery layer that resembles Google's business listings.

OpenAI上线Business Hub:ChatGPT开始做商家资料与商业智能体
Read details: Anthropic launches Claude Meetings to turn workplace discussions into plans and tasks
AI News / Anthropic / Claude

Anthropic launches Claude Meetings to turn workplace discussions into plans and tasks

Anthropic launches Claude Meetings to turn workplace discussions into plans and tasks. As participants talk, Claude can draft the plan being discussed, map out a timeline on a whiteboard, answer questions in the meeting chat, and access connected work apps to retrieve information, such as last year's figures.

AI动态:Anthropic launches Claude Meetings to turn workplace discussions into plans and tasks
Read details: RainFocus Launches Anthropic Connector Giving AI Agents Access to Event Data
Anthropic / Agent / Enterprise

RainFocus Launches Anthropic Connector Giving AI Agents Access to Event Data

RainFocus has launched an Anthropic connector that gives AI agents access to its event data, StreetInsider reported. The launch extends Anthropic’s agent ecosystem into vertical industry systems, though the report leaves the connector’s capability limits, ship date and data governance model undisclosed.

Read details: NVIDIA Expands Open Source Commitment With OpenShell Donation and New AI Infrastructure
AI News / NVIDIA / Infrastructure

NVIDIA Expands Open Source Commitment With OpenShell Donation and New AI Infrastructure

NVIDIA Expands Open Source Commitment With OpenShell Donation and New AI Infrastructure. Kubernetes, the open source system that manages how applications run across thousands of AI factories, has become the foundation for AI, and the demands on it are only growing. To help meet them, NVIDIA is deepening its commitment to open source at KubeCon + C...

AI动态:NVIDIA Expands Open Source Commitment With OpenShell Donation and New AI Infrastructure
Read details: MiniMax posts H3 image model on Hugging Face with a bundled prompt expander
MiniMax / Image Model / Open Source

MiniMax posts H3 image model on Hugging Face with a bundled prompt expander

MiniMax has published MiniMax-H3-Image on Hugging Face, a text-to-image repository whose tags also cover image-to-image and editing workflows. Alongside it, a companion prompt-expansion model called MiniMax-H3-Image-PE ships with fp4 tagging, hinting at low-precision inference.

MiniMax 上线 H3 图像模型:文生图与图像编辑同框进 Hugging Face
Read details: Anthropic sets IPO range of $70 to $78 a share, implying a $1.404 trillion valuation
Anthropic / IPO / AI Business

Anthropic sets IPO range of $70 to $78 a share, implying a $1.404 trillion valuation

Anthropic has set an IPO price range of $70 to $78 per share, planning to sell 825 million Class A shares in a sale that could raise up to $64.35 billion. At the top of the range, the Claude maker would be valued at $1.404 trillion on a fully diluted basis, well above its $965 billion valuation in May.

Anthropic IPO定价区间70至78美元 隐含估值最高1.404万亿美元
Read details: Google Cloud launches Gemini for healthcare to unify clinical data and automate workflows
Google Cloud / Gemini / Healthcare

Google Cloud launches Gemini for healthcare to unify clinical data and automate workflows

Google Cloud has launched Gemini for healthcare, aiming to unify clinical data and automate workflows for hospitals and care providers. The company is pitching the product around reducing administrative burnout, marking another push by a major cloud vendor into vertical AI solutions.

谷歌云推出医疗版Gemini,整合临床数据并自动化工作流
Read details: Cyberattacks threaten to make open source AI a little less open, Hugging Face exec warns
AI News / Hugging Face / Business

Cyberattacks threaten to make open source AI a little less open, Hugging Face exec warns

Cyberattacks threaten to make open source AI a little less open, Hugging Face exec warns. Hugging Face executive Irene Solaiman warns that growing cyberattacks on open AI platforms could force tighter security restrictions, threatening the accessibility that makes open source communities valuable to developers worldwide.

AI动态:Cyberattacks threaten to make open source AI a little less open, Hugging Face exec warns
Read details: OpenAI Academy Brings Free ChatGPT Workshops to Caregivers in Five U.S. Cities
OpenAI / ChatGPT / AI Literacy

OpenAI Academy Brings Free ChatGPT Workshops to Caregivers in Five U.S. Cities

OpenAI Academy is bringing free, hands-on ChatGPT workshops to caregivers in five U.S. cities, aiming to help them build practical AI skills for everyday life. The program targets a group that often lacks the time and tailored guidance to adopt new tools on its own.

Read details: Nvidia unveils GR00T 2 world-action model, with models, data and simulation to be open-sourced
NVIDIA / Robotics / Open Source

Nvidia unveils GR00T 2 world-action model, with models, data and simulation to be open-sourced

Nvidia's GEAR team introduced GR00T 2 at CoRL in Austin, a world-action model that jointly predicts future frames and robot actions, with 100,000 hours of human data already released. Yuke Zhu said the model, data, simulation environments and capture hardware will be open-sourced, backed by a toolchain spanning data collection to real-robot deployment.

Read details: Hanqiao, an AI agent for international Chinese education, launches at BLCU
Education / Agent / China

Hanqiao, an AI agent for international Chinese education, launches at BLCU

Beijing Language and Culture University reported on Nov. 10 that a launch event for Hanqiao, an AI agent for international Chinese-language education, was held on campus. The university's post discloses only that the launch took place, with no base model, feature list or availability date.

Read details: OpenAI report finds 79% of students use AI for homework, but only 47% were shown how
OpenAI / Education / Policy

OpenAI report finds 79% of students use AI for homework, but only 47% were shown how

OpenAI published a report on the education guidance gap, drawing on its EMEA partnerships in Estonia, Slovakia, Greece, and Italy and on new ThinkYoung research. The survey found 79% of students use AI for assignments while only 47% said anyone had shown them how.

Read details: Flexxon wins third CyberCall grant to build hardware-level security for AI agents
Flexxon / AI Security / Hardware

Flexxon wins third CyberCall grant to build hardware-level security for AI agents

Flexxon has been named a CyberCall 2025 awardee under the 'Cybersecurity for AI' challenge and will develop X-PHY AI Armor, a hardware-rooted secure MCP gateway. The project sits between AI agents and the systems they reach, moving authorization and audit records for high-risk actions into hardware.

Read details: Singapore Airlines doubles Kris chatbot satisfaction after moving to OpenAI models
OpenAI / Singapore Airlines / Enterprise AI

Singapore Airlines doubles Kris chatbot satisfaction after moving to OpenAI models

Singapore Airlines moved its Kris customer service chatbot from scripted dialog flows to an LLM-powered experience built on the OpenAI API and saw customer satisfaction double. The airline also uses OpenAI models in its Flight Recommender, its employee platform Jarvis and its software development workflow, with agent handling time down about 10%.

Read details: openJiuwen debuts TTSE, a two-track experience framework for agents
Agent / Open Source / LLM

openJiuwen debuts TTSE, a two-track experience framework for agents

openJiuwen has released TTSE, an experience-learning framework that splits agent knowledge into FACT entries about the environment and TIP entries about task-specific execution methods, each with its own numbering and evidence trail. The team reports gains over single-track baselines on ALFWorld, ScienceWorld, GDPevo, SOPBench and PinchBench.

Read details: Ant's Lingbo open-sources LingBot-Touch, a general sense of touch for robots
Robotics / Embodied AI / Open Source

Ant's Lingbo open-sources LingBot-Touch, a general sense of touch for robots

Ant Group's embodied AI unit Lingbo has open-sourced LingBot-Touch, a tactile foundation model that transfers touch experience from human hands to robot hands absent from its pretraining, with no parameter changes when attached to a new XHand. QbitAI reports π0.5's average success rate rising from 48.00% to 74.17% across six tasks.

Daily Archive

Daily Archive

Full Archive
2026-11-18Daily AI Archive | 2026-11-18This issue collects 50 AI updates from 2026-11-18, led by Google makes Tasks in Gemini free in six countries and deepens Workspace integration, Google to keep Gemini chats for two months, up from 72 hours, even with history off, Voyage Labs raises $12M to build Atlas, a personal AI agent for experiencing the world, Daydream's New AI Agent Acts as a Personal Stylist — and Shops for You.2026-11-17Daily AI Archive | 2026-11-17This issue collects 40 AI updates from 2026-11-17, led by Lambda Adopts NVIDIA Photonics CPO Switches to Harden 10,000 GPU AI Factories, Report: OpenAI Explored Buying Interpretability Startup Goodfire for Over $5 Billion, NVIDIA Extends Slurm to Put Quantum Processors Inside Supercomputing Workflows, Nvidia's Vera Rubin debuts in MLPerf with a training sweep and 10x agentic efficiency gain.2026-11-16Daily AI Archive | 2026-11-16This issue collects 35 AI updates from 2026-11-16, led by ByteDance opens cheap Seedance 2.5 Flash API, with a 720p ceiling, XRP Ledger logs 11.7 million sub dollar AI agent payments, t54.ai bets on settlement, OpenAI Launches ChatGPT Wallet for AI Agent Payments, 腾讯混元字节研究人员集结!北京AI智能体大会及个人Agent研讨会要来了 智东西.2026-11-15Daily AI Archive | 2026-11-15This issue collects 22 AI updates from 2026-11-15, led by OpenAI will let ChatGPT Health conversations feed model training from November 19, Asian governments move to lock in Nvidia chip supply after Jensen Huang's exit, DeepSeek Opens Chatbot Search, File and Voice Features to Developers, AWS Trainium Picks Up SGLang Support: 17 Models, Kimi K3 on Three Nodes.2026-11-14Daily AI Archive | 2026-11-14This issue collects 21 AI updates from 2026-11-14, led by ChatGPT adds a Wallet so AI agents can pay at checkout, Gemini 4 Hydrogen costs 76% less per task than 3.8 Flash, LexisNexis launches an AI agent detection tool, Cisco CEO says every employee inside the company will get a personal AI agent.2026-11-13Daily AI Archive | 2026-11-13This issue collects 36 AI updates from 2026-11-13, led by AMD Unveils New AI Data Center GPU, Shares Jump More Than 5%, After Huang's resignation, Trump floats Tim Cook to run Nvidia, Northern Trust and Microsoft deploy an AI agent that cuts onboarding prep from weeks to days, Bielik 11B v3.1 ships as a 56 language Polish open model.2026-11-12Daily AI Archive | 2026-11-12This issue collects 45 AI updates from 2026-11-12, led by OpenAI's quarterly revenue reportedly tops $10B for the first time, Simpro Launches JustDo AI Agent for Trade Businesses, Gopuff goes live inside Grok Bot with a fully integrated checkout, TechRadar Reports Grok Build Coding Agent Is Leaking Confidential Data.2026-11-11Daily AI Archive | 2026-11-11This issue collects 31 AI updates from 2026-11-11, led by Google Expands Its TPU Accelerator Family, Framed Around the AI Agent Era, RainFocus Launches Anthropic Connector Giving AI Agents Access to Event Data, NVIDIA Expands Open Source Commitment With OpenShell Donation and New AI Infrastructure, Nvidia CEO Jensen Huang’s affair with alleged Chinese spy: a timeline.2026-11-10Daily AI Archive | 2026-11-10This issue collects 39 AI updates from 2026-11-10, led by Northern Trust and Microsoft Deploy AI Agent, Cutting Client Onboarding Time by 83%, Anthropic Buys $3.4B of Natural Gas Turbines as AI Power Race Widens, Google Adds TPU 8tx and 8i Max to Its Gen 8 Lineup, Aimed at Agent Workloads, Medium's New TK App Puts AI Assistance Inside the Writing Process.2026-11-09Daily AI Archive | 2026-11-09This issue collects 29 AI updates from 2026-11-09, led by Cloudflare and Tools for Humanity Propose Identity Framework for AI Agents, Nvidia board opens independent review of Jensen Huang's ties to alleged Chinese spy, Report Says Anthropic Is Building AI Systems Twice as Fast as Last Year, GammaDX index finds only 12 of 50 Australian websites are agent ready.