Skip to content
K = 0.731
All news · Today70 items

All AI news

All collected and published items, ordered by time. Picks are a selected subset.
10/12Mon
  1. The Verge · AI75

    Microsoft announces Maia 300 AI chip to reduce reliance on Nvidia GPUs

    Microsoft announced Maia 300, its third-generation in-house AI chip, with production rollout starting in 2027 and plans to make Maia its primary AI inference system. Microsoft expects 40 percent more tokens per megawatt than leading GPU-based systems and up to 7x Maia 200’s rack-scale token throughput in a full 96-accelerator configuration. Built on TSMC’s N2P process node, Maia 300 separates compute, memory controller, and I/O functions through a new chiplet architecture.

  2. Satya Nadella · X73

    Microsoft introduces Maia 300 and plans multi-gigawatt Maia deployments

    Microsoft is introducing Maia 300, which Satya Nadella says delivers 40% more tokens per watt compared to leading GPU-based systems. Starting next calendar year, Microsoft will scale its Maia family to multi-gigawatt deployments, making it the primary inference system for its own products and services. Microsoft is already deploying Maia 200 across the US, Europe, and Australia, delivering up to 1.7x more tokens per watt across experiences in Teams and Copilot, with a goal of building on full-stack efficiency gains while continuing to bring ecosystem innovations into its fleet.

    Why it matters: Making Maia the primary inference system for Microsoft’s own products and services gives its stated efficiency gains a concrete deployment context.

  3. Qwen · X61

    Qwen introduces QwenGyre for elastic RL in extreme-long-horizon agent training

    Qwen introduces QwenGyre, bringing elastic RL to extreme-long-horizon (XLong) agent training. Its Elastic Scheduler reallocates GPUs between rollout and training without interrupting live executions, while its Trajectory Processor converts branching executions into prefix-shared trajectory trees with bounded, deduplicated training samples. Scaled to Qwen3.8-2.4T-A95B with ~700K tokens per rollout, QwenGyre raises NL2RepoBench from 52.5% to 58.5% in 48 steps and cuts end-to-end training time by ~44% vs Async. Paper: https://arxiv.org/abs/2609.33848

  4. Import AI34

    Import AI 476: New Benchmarks Test Visual Reasoning and Research Taste

    Import AI 476 covers Humanity’s Sixth Sense, a benchmark from Elorian and Scale AI that tests intuitive visual reasoning. People scored 93.1 aggregate pass@1, while GPT 6 Astra scored 53.6; GLM-5.3-flash and Kimi K3 scored 26.8 and 25.5. The newsletter also examines how AI could accelerate space development and reports on TasteVal, a benchmark of AI systems’ research taste.

  5. The Decoder59

    Andrej Karpathy describes a widening gap in how people experience AI capabilities

    Andrej Karpathy describes four tiers of LLM exposure, arguing that public perceptions lag behind capabilities experienced by frontier-system users. He estimates roughly six billion people have barely encountered LLMs, one to two billion use casual chatbot products, 20 million use frontier models professionally in math and code, and 5,000 access internal frontier systems. He claims those internal systems coordinate thousands of agents on projects that would take experts years, while human review falls behind, and attributes progress in coding and math to reinforcement learning on verifiable rewards and commercial training priorities.

  6. The Decoder70

    White House restores UK AI Security Institute's access to frontier models

    The White House has lifted its block on the UK's AI Security Institute accessing new models from Anthropic, Google DeepMind, and OpenAI, The Daily Telegraph reports. The reversal followed diplomatic pressure from AI Minister Kanishka Narayan after September's restrictions led Anthropic to restrict access to Claude Mythos 5.1. Narayan plans to announce new safety measures, including stricter reporting requirements for AI-related security incidents involving critical infrastructure.

  7. TechCrunch · AI57

    Former Plaud exec raises $26M for Ambi, an AI app that listens in the background

    Former Plaud executive Peter Mo’s startup Ambi has raised $26 million across multiple funding rounds led by Luminous Ventures and Huaqin. It is releasing apps for iPhones, Macs and Apple Watch that can capture conversations throughout the day, with an AI agent intended to offer suggestions and begin tasks during meetings. Mo says ambient capture is off by default, but users remain responsible for informing others; Ambi says it stores no recordings and encrypts data without specifying the encryption method.

  8. 雷锋网62

    How Anta Uses Volcano Engine Agents Across Global E-Commerce Workflows

    Anta is using Volcano Engine Agents for promotional videos, overseas ordering, and cross-border returns. The article reports that an ArkClaw workflow packages four ordering templates into Skills, reducing ordering decisions from half a day to 10–20 minutes, cutting manual entry by about 90%, and achieving data accuracy above 99%. A returns Agent automatically creates over 1000 return documents monthly for Anta’s TikTok cross-border business, with expansion to FILA and other brands planned.

  9. 雷锋网22

    Huawei’s Eight Years of Foldable Phones Put Its Mate X7 Ahead of Apple’s First Foldable

    Huawei’s Mate X7 continues an eight-year evolution of foldable designs, while Apple’s iPhone Duo has just entered the market. IDC data for the first quarter of 2026 puts Huawei first in China’s foldable phone market with a 60% share. The article highlights Huawei’s range of foldable formats, the Mate X7’s multitasking features and hinge design, and its 11,999-yuan starting price, below the iPhone Duo’s.

  10. The Decoder81

    Google DeepMind’s AlphaProof Nexus solves dozens of open math problems in peer-reviewed study

    Google DeepMind’s AlphaProof Nexus solved dozens of open math problems, according to a peer-reviewed study in Science, including nine of 353 open Erdős problems, two unsolved for 56 years. It also proved 44 OEIS conjectures, solved an open algebraic geometry question, and improved a min-max optimization bound. Built around Gemini 3.1 Pro, it drafts and revises proofs in Lean using checker feedback and can call AlphaProof for harder sub-problems, though DeepMind reports it solved the nine Erdős problems without that tool.

    Why it matters: DeepMind’s use of peer review provides a concrete comparison with OpenAI’s publication of AI proofs on GitHub amid a debate over how AI mathematics should be evaluated.

  11. 雷锋网48

    Geely Launches Galaxy Battleship 700 SUV at 169,800 Yuan

    Geely launched its first new-energy rugged SUV, the Galaxy Battleship 700, on October 10 in Hangzhou, with three powertrain options and six trims. Its limited-time launch price starts at 169,800 yuan. Officially stated specs include a maximum combined range of 1,770 km and, for the 2.0T Summit trim, 0-to-100 km/h acceleration in 3.95 seconds. The article notes that prices and performance figures vary by trim.

  12. 雷锋网47

    Orbbec Ends Robotic Lawn Mower ODM Business After Shipments Reach Only Thousands

    Orbbec decided in mid-September to end its robotic lawn mower ODM business after cumulative shipments reached only a few thousand units, far below expectations. Its vision module business remains unaffected, and perception modules for outdoor robots are still in development. The report cites tighter FCC rules, a crowded market and customer shipments too small to sustain the ODM product line.

  13. 雷锋网16

    Baidu World 2026 to Be Held in Beijing on November 12

    Baidu announced that Baidu World 2026 will take place on November 12 at the China National Convention Center Phase II in Beijing, under the theme “The Next Act: Intelligent Organizations.” The event plans one main forum, 12 breakout forums, more than 30 AI classes, and a 6,000-square-meter exhibition area. Its forums will cover AI-driven organizational change, including AI office tools and enterprise deployment of agents.

  14. 量子位47

    Zhipu’s GLM-5.3-Highspeed Model Reportedly Reaches 490 Tokens per Second

    Zhipu has added the GLM-5.3-Highspeed model to Zcode, where it is being gradually rolled out to subscribers at random. A user reported a generation speed of 490 tokens per second; the article says the model is currently available for testing in Zcode 3.15.1. The reported speed has not been independently confirmed, and the author says they could not access the model.

  15. 量子位75

    Huawei and UCL’s Memento 3 Clears 25 ARC-AGI-3 Games Without Updating Model Weights

    Huawei’s Noah’s Ark Lab, UCL, and collaborators report that Memento 3 cleared all 25 public ARC-AGI-3 games and 183 levels using 7518 actions, about 56% fewer than humans, achieving a 100 RHAE score. Built on Claude Opus 5 without weight updates or game-specific fine-tuning, the agent iteratively revises a Markdown rulebook and executable Python world model. In Atari Pong, its learned code-based strategy won three games 21:0 without further LLM calls; the experiments primarily cover game environments with relatively clear rules.

  17. 雷锋网9

    Claude Banned 11.4 Million Accounts in the First Half of 2026, With 42,000 Appeals Reversed

    Claude banned 11.4 million accounts in the first half of 2026, according to an Anthropic report published in July that was recently recirculated. Of 398,000 appeals submitted, 42,000 were successful. Anthropic did not disclose a full breakdown of the reasons for bans, but the report listed several types of activity that may trigger them.

  18. 雷锋网57

    Microsoft releases Qwen-based Microsoft-Decision-1 and tests applications across domains

    Microsoft released Microsoft-Decision-1, a decision model post-trained from Qwen3.5-9B, and says it outperforms LLMs and decision models including Jev 1.13.0 on latency and accuracy in structured decision tasks. Its Xbox research team has used it to process more than 10,000 player feedback entries, while CEO Satya Nadella says Microsoft is testing applications across domains, including incident response, quality control, and scientific discovery. The report describes decision models as smaller, faster models that output predictions, probabilities, and confidence scores rather than natural language.

  20. 宝玉 · X26

    Bao Yu Wonders Whether He Fits AI Startups’ Hardest-to-Hire Role

    Bao Yu describes a role AI startups reportedly find difficult to hire for: someone who follows online trends, creates engaging content, understands AI and technology, and executes well. He says content creators, technical evangelists and engineers, and traditional PR staff each tend to lack some of those abilities, while few people have all four.

  21. 宝玉 · X58

    Anthropic’s Claude Subscription Rules Appear Looser, but Amp’s Status Remains Unclear

    Baoyu says Anthropic appears to have relaxed restrictions on Claude Agent SDK and claude -p usage, but has made no public statement specifically addressing Amp or guaranteeing against account bans. Quoted author Quinn Slack interprets a policy update from 2 days ago and Anthropic employee comments as allowing SDK and third-party apps to use subscription allowances, citing https://support.claude.com/en/articles/15036540-use-the-claude-agent-sdk-with-your-claude-plan and https://news.ycombinator.com/item?id=49997143. Baoyu recounts earlier restrictions and says Anthropic paused its planned June 15 switch to separate billing for Agent SDK and claude -p.

  23. Ethan Mollick · X40

    GPT-6 Pro finds supposed errors in academic papers all the time, just as part of other work. Confirming the errors & learning their importance would still take a lot of time today, but it will get faster. We don't have the right mechanisms in place to deal a flood of corrections.

    GPT-6 Pro finds supposed errors in academic papers all the time, just as part of other work. Confirming the errors & learning their importance would still take a lot of time today, but it will get faster. We don't have the right mechanisms in place to deal a flood of corrections.

  24. Ethan Mollick · X23

    It has become disheartening that the average replybot on this site has shifted its replyslop angle to asking interesting questions. It feels like actual intellectual engagement, which makes it somehow worse than just rephrasing the post with a "that's the part people miss" bit.

    It has become disheartening that the average replybot on this site has shifted its replyslop angle to asking interesting questions. It feels like actual intellectual engagement, which makes it somehow worse than just rephrasing the post with a "that's the part people miss" bit.

  25. 雷锋网38

    Former DeepRoute AI Executive Liu Nianqiu Says Mapless Driving Experience Can Transfer to Robots

    Femo One founder Liu Nianqiu says experience moving DeepRoute AI from high-definition maps to mapless driving can inform the company’s approach to robots, which he calls “Physical Agents.” Femo One is focused on open, dynamic environments and plans to begin deployments at multiple sites next year, building a cycle of deployment, data collection, training and redeployment.

  26. 宝玉 · X51

    BaoCut Adds Claude Design and Claude Slides Alongside Video Editing

    Baoyu integrated reverse-engineered Claude Design and Claude Slides into BaoCut, an open-source Agent originally focused on video editing. He reports substantially improved design performance when GPT-6.1 Sol uses the Claude Design Skill, possibly because Claude Design defaults to Claude Design System Skill; standalone Skills would not support editing and annotation, so he integrated them into BaoCut. Users can download or run BaoCut locally, use a local agent CLI such as codex, and select the design skill during a session. https://github.com/JimLiu/baocut/commit/7aa049e8 https://baocut.com

  27. 宝玉 · X57

    Satya Nadella argues enterprises should treat AI models as insider risks

    Baoyu summarizes Microsoft CEO Satya Nadella’s argument that enterprises should manage AI models as insider risks, with identities, restricted permissions, audit records, and controls to halt operations. Nadella outlines seven principles covering model diversity, observability, verification, independent control and auditing, containment, and incident disclosure. He calls for disclosing chains of thought while acknowledging that they may not faithfully reflect model reasoning, and argues that security controls must operate outside the model.

  28. Ethan Mollick · X32

    I spoke to some annoyed leaders at big firms who were still big on GPTs as of last month, before GPTs were sunsetted. Yes, GPTs outlived their usefulness. But large organizations have complex planning & all the Labs (not just OpenAI) keep making sudden changes that confuse them.

    I spoke to some annoyed leaders at big firms who were still big on GPTs as of last month, before GPTs were sunsetted. Yes, GPTs outlived their usefulness. But large organizations have complex planning & all the Labs (not just OpenAI) keep making sudden changes that confuse them.

  29. 宝玉 · X61

    Baoyu uses REA and Computer Use to investigate CapCut subtitle templates

    Baoyu describes using REA with Computer Use to investigate how CapCut renders dynamic subtitles, without knowing where their formats were stored. He had it operate CapCut, download Templates, and monitor network URLs and disk save locations. After downloading several dozen through clicks, Codex found clicking too slow and constructed URLs to download directly, ultimately downloading 2291 Templates.

  31. Ethan Mollick · X57

    Ethan Mollick argues advanced model use at labs remains concentrated in a few fields

    Ethan Mollick argues that many of the 5,000 people at labs with access to advanced models still haven’t used them to their full extent, with use concentrated in cybersecurity, math, and AI R&D. There are many other fields to advance and problems to solve, he adds, responding to @karpathy’s argument that the gap in shared understanding of LLM capabilities is widening.

  32. 宝玉 · X62

    Using REA to Trace How Codex Attaches Browser Annotations

    Baoyu describes using the open-source REA (Reverse Engineer Anything) MCP service to understand how Codex and the ChatGPT app attach browser annotations to the composer, without decompiling Codex itself. After REA traced the Electron app's code path, Codex built a test page that could open a comment box and prefill input via document.oai.annotation.request(), an OpenAI API the author had not known about. Cursor and Claude Code did not support a similar API.

  33. 宝玉 · X22

    Toyota's Kaizen Rule Applied to AI: Don't Cut Staff After Efficiency Gains

    A Japanese blogger cited Toyota's Kaizen rule — never cut staff after process improvements reduce work hours — and argued the same applies to AI adoption. Companies that cut headcount as soon as AI boosts efficiency will find employees hide their best AI methods, he said. A quoted post adds that DORA research found organizations addressing employee job concerns had higher team AI adoption.

  34. TechCrunch · AI40

    Apple to Announce HomeView Smart Home Hub on Tuesday, Gurman Reports

    Apple will announce a new smart home hub called the HomeView on Tuesday, according to Bloomberg's Mark Gurman. Gurman describes it as a Siri-powered device with a display and an A18 chip for controlling other smart home devices, reportedly available in countertop and wall-mounted models. Apple will also unveil updated Apple TV and HomePod mini devices, and Gurman said earlier this week that Apple plans to announce an LG partnership for a smart lock, thermostat, and doorbell.

  35. 宝玉 · X62

    Google pilots AI-assisted coding interviews for software engineers

    Google is piloting a new software engineer interview round called Code Comprehension, in which candidates read, debug and optimize existing code with Gemini open, Business Insider reported in May based on an internal document. A Google spokesperson confirmed Gemini is the pilot assistant. The round targets junior to mid-level roles in select US teams, starting with Google Cloud and Platforms and Devices, and assesses prompt writing, output checking and debugging.

  36. 宝玉 · X62

    Karpathy Says the LLM Capability Gap Is Becoming a Funnel

    Andrej Karpathy argues the gap in shared understanding of LLM capability is widening into a sharp funnel: roughly 6B people have barely touched LLMs, 1-2B use free-tier chatbots casually, about 20M use frontier models for math and code, and roughly 5,000 have internal access to frontier systems. He attributes the funnel to problem ambition, capability jaggedness in verifiable domains, and access tiers.

  38. The Verge · AI38

    Silicon Valley leaders adopt Trump's "super intelligence" rebrand for AI

    Some of Silicon Valley's most powerful figures are falling in line with President Donald Trump's push to rebrand artificial intelligence as "super intelligence." Elon Musk promised to rename SpaceXAI to SpaceXSI, Marc Benioff said Salesforce would rename AIForce to SIForce, Nvidia tied "super intelligence research" to a $1 billion commitment, and Microsoft's Satya Nadella repeatedly used the term in a post calling for an AI "emergency brake."

  39. TechCrunch · AI8

    TechCrunch Disrupt 2026 Opens in 48 Hours in San Francisco

    TechCrunch Disrupt 2026 opens in less than 48 hours at San Francisco's Moscone West, running October 13-15. More than 10,000 founders, investors, and tech leaders are expected, alongside 250+ speakers across 200+ sessions on six stages and 300+ startups exhibiting. Speakers include RJ Scaringe of Rivian, Mark Wahlberg, Alexa von Tobel of Inspired Capital, Andrew Feldman of Cerebras Systems, Cat de Jong of Anthropic, and Les Karpas of NVIDIA. Tickets cost up to $100 less before door prices take effect October 13 at 8 a.m. PT.

  40. Andrej Karpathy · X34

    Karpathy: LLM Capability Gap Is Widening Into a Sharp Funnel

    Andrej Karpathy says the gap in shared understanding of LLM capability is widening into a "sharp funnel" rather than two groups talking past each other. He estimates roughly 6B people have barely encountered LLMs, 1-2B use free-tier ChatGPT-like products casually, about 20M see frontier-grade math and code use, and roughly 5,000 have internal access to frontier systems. He attributes the funnel to impact scaling with ambition, the jaggedness of capability, and access tiers.