Skip to content
K = 0.731
All news · Today68 items

All AI news

All collected and published items, ordered by time. Picks are a selected subset.
10/14Wed
  1. 量子位60

    APUS AI Open-Sources Reflex Agent With Model-Driven Control and Persistent Memory

    APUS AI announced its MIT-licensed Personal Agent, Reflex Agent, on October 13, offering desktop, Web, and SDK access. Its Reflex decision model handles routing, skill selection, result review, and execution supervision, while Continuum retrieves historical information to support long-running tasks without exceeding individual model calls’ input limits. In material supplied by APUS AI and republished by QbitAI, the company reports responses up to 3 times as fast as Claude Code and Codex in tested scenarios; the agent is at https://github.com/APUS-AI-Lab/APUS-Reflex-Agent and the decision model at https://huggingface.co/apus-ailab/APUS-Reflex.

  2. 量子位45

    SenseTime’s Three Agents Bring AI to Shanghai’s “Quantum City” Governance

    SenseTime’s AI platform, based on the upgraded “Cloud Universe Starry Sky” model, developed and delivered three agents for Shanghai’s first batch of government agents under its “Quantum City” initiative. They handle flood-control dispatch, spatial analysis, and planning updates. In the Suzhou Creek basin, the “Yuzhi Heluo” agent improved dispatch efficiency by 80%, alongside reported gains of 65% in spatial computing and 70% in dynamic simulation.

  3. 量子位53

    PHANES AI and Tianlong Mobile Launch SeePhys Hub for Whole-Body Data Collection

    PHANES AI and Tianlong Mobile launched SeePhys Hub on October 14 to supplement first-person human demonstration data with whole-body motion and optional tactile pressure recordings. The system combines a four-camera headband, two wristbands and three waist and leg trackers, with software for processing data and mapping human movements to robots. PHANES AI plans to release a whole-body, all-modality world model in 2027 and prioritize application validation in industrial, restaurant and retail settings.

  9. Alexandr Wang · X26

    Alexandr Wang Urges Muse Wonder to Invest in Its Tech Tree

    @Hesamation is benchmarking models on Stronghold Crusader, a strategic game they grew up with. They said Muse Wonder playing the game on an old laptop is “ridiculously fun to watch,” and that they have tested Opus, Sol, GLM flash, Kimi K3, and more models. Alexandr Wang reacted: “cmon muse wonder i believe in you invest in your tech treeeeeeee” https://t.co/dSUEmGVxV7

  10. 雷锋网74

    Jiayuan Mao explains single-demonstration robot learning and 83% real-world generalization at IROS 2026

    Reporting on Jiayuan Mao’s IROS 2026 presentation, Leiphone describes research from UPenn, MIT, and Stanford on robot learning and compositional generalization. RAPID combines a coding agent with physics validation to learn complex non-grasping tasks from one human demonstration. In separate video-action research, the team introduced Temporal Ratio to measure attention to future versus current frames; adaptive guidance based on this metric raised average success on unseen compositional tasks with a real dual-arm robot from 71.8% to 83% without retraining.

  11. François Chollet · X31

    The ideal outcome for science isn't a couple of labs using proprietary models and millions of dollars to crack trophy problems, it's to have millions of AI-aided human scientists who follow their curiosity and become way more ambitious

    The ideal outcome for science isn't a couple of labs using proprietary models and millions of dollars to crack trophy problems, it's to have millions of AI-aided human scientists who follow their curiosity and become way more ambitious

  13. 雷锋网51

    Tencent WorkBuddy Enterprise launches an observability center in public beta

    Tencent WorkBuddy Enterprise added an observability center to its admin console on October 13 and opened it for public beta. It provides real-time monitoring, session tracing, and execution trace details covering usage, credit consumption, response times, and errors. Administrators can break down consumption by department and model, investigate specific execution nodes, and adjust Agent prompts, models, and tools at https://www.workbuddy.cn/admin/overview.

  15. Google DeepMind · X40

    Over half of reported crimes in Singapore use sophisticated social engineering. With @GovTechSG, we’re building on-device fraud detection using our open model Gemma. By fine-tuning it on local scam tactics, it identifies threats in real time without compromising personal privacy → https://goo.gle/4ygVkJa

    Over half of reported crimes in Singapore use sophisticated social engineering. With @GovTechSG, we’re building on-device fraud detection using our open model Gemma. By fine-tuning it on local scam tactics, it identifies threats in real time without compromising personal privacy → https://goo.gle/4ygVkJa

  20. Google DeepMind63

    Google DeepMind and Singapore GovTech deliver on-device Gemma model for scam detection

    Google DeepMind and Singapore’s Government Technology Agency (GovTech) have delivered a Gemma model fine-tuned on local scam conversations and optimized for typical smartphones in Singapore. A lightweight classifier screens incoming messages before the model evaluates broader conversation context and triggers warnings about suspected manipulation, with conversation processing kept on the device. Training excluded private or personally identifiable information, using anonymized datasets derived from scam reports and intelligence feeds.

    Why it matters: The two-stage detection pipeline offers a transferable approach to balancing smartphone power consumption with contextual scam analysis while keeping private conversations on the device.

  24. Ethan Mollick · X29

    Argon Creates a Game About Imminence

    Argon is quite different: “A game about Imminence. Something very big, very strange is happening. A suburb & the arrival of a vast & unknowable presence. Not horror, invoke the feeling of the end of all things coming, inevitably, but also not sad or scary” https://t.co/9hiPnuqexe https://t.co/hzJ3lAi0ez

  25. Ethan Mollick · X15

    Hey, AI give me your best "Unless it is from X, its just sparkling Y" style joke": GPT-6 Pro: "Unless it comes from the Stockholm region of Sweden, it’s just sparkling employee engagement." Gemini 4 Argon High: "Unless it comes from the 3:00 AM region of staring at the ceiling,

    Hey, AI give me your best "Unless it is from X, its just sparkling Y" style joke": GPT-6 Pro: "Unless it comes from the Stockholm region of Sweden, it’s just sparkling employee engagement." Gemini 4 Argon High: "Unless it comes from the 3:00 AM region of staring at the ceiling,

  26. Artificial Analysis · X46

    Claude Sonnet 5.5 Scores 68 on the Coding Agent Index at Max Effort

    In Claude Code, Sonnet 5.5 scores 68 on the Coding Agent Index at max effort and 63 at xhigh. The xhigh setting costs $3.33 per task versus $14.19 at max, a 77% reduction for 5.5 fewer Index points. Max generates 4.8× as many output tokens and takes 3.4× as many turns, while improving all three benchmark scores; its largest gain is on Terminal-Bench 4.0, from 58.1% to 66.2%.

  27. Ethan Mollick · X37

    We will soon see what will be, in effect, a perfectly legal distributed denial of service attack on customer service departments authorized by millions of consumers when it becomes widely known that Muse/Dot/etc can make unlimited requests for old refunds, negotiate deals, etc.

    We will soon see what will be, in effect, a perfectly legal distributed denial of service attack on customer service departments authorized by millions of consumers when it becomes widely known that Muse/Dot/etc can make unlimited requests for old refunds, negotiate deals, etc.

  28. 宝玉 · X81

    Why Anthropic built Scion to replace git for its monorepo

    Drawing on Anthropic’s engineering blog, Baoyu explains why the company replaced git with Scion as daily merged code changes grew from around 2,500 in May to more than 30,000 in September. Scion retrieves files on demand, moves operations to the server, and uses layered caches with hit rates above 99%. Anthropic reports that median agent workspace startup fell from 7.1 minutes to 5.1 seconds, while emphasizing that Scion serves exceptionally large monorepos and is not a general-purpose git replacement.

    Why it matters: The comparison illustrates how assuming continuous server access changes source control architecture, connecting on-demand file retrieval and server-side operations to the infrastructure demands of agent-generated code.

  31. Simon Willison47

    Profiling web-edit with Python 3.15 Reveals Line-Ending Workload Hotspots

    Profiling the unmodified web-edit app with Python 3.15’s built-in `profiling.sampling` found that preserving mixed line endings dominated the editing workload. `preserve_endings` appeared in 64.8% of 15,296 CPU stacks, while `SequenceMatcher.find_longest_match` contributed 38.3% of leaf samples; 95.1% of stacks came from AnyIO’s worker thread, making all-thread sampling essential.

  32. Simon Willison64

    Simon Willison updates llm-meta-ai 0.2 for Meta AI’s Muse Wonder

    Simon Willison updated his llm-meta-ai plugin to support Meta AI’s newly released Muse Wonder models. He shares pelican drawings across six reasoning levels, from minimal to max, and compares them with Muse 1.3 and Muse 1.2, reporting material drawing improvements between releases. He suggests these may be Meta’s first Fable class models and reports that Artificial Analysis scores Muse Wonder (Max) as equivalent to GPT-6.1 Sol.

  34. Simon Willison74

    Simon Willison adds Gemini 4 Argon support to llm-gemini 0.35

    Simon Willison added Gemini 4 Argon support to llm-gemini, following the model’s availability today through the Gemini API and a $100/month subscription plan. He praised its SVG pelican illustrations, with the high illustration costing 17.166 cents. Argon launches at $2/million input and $10/million output, with prices set to double on December 1st; Artificial Analysis reports $2,407 to run its benchmark, which Willison considers a more useful comparison than per-token pricing.

  35. Simon Willison76

    How Anthropic redesigned source control with Scion

    Simon Willison highlights Anthropic’s explanation of Scion, its custom alternative to Git, as merged code changes grew from around 2,500 per day in May 2026 to over 30,000 per day in September. Anthropic reports almost 100 GB of main-branch history and roughly 25 million daily CI checkouts; seven people built the system, completing migration in five months. Willison speculates that Fable/Mythos class models contributed to the increased coding velocity and connects Anthropic’s experience with adversarial review to his own, including the need for engineers to steer models away from over-engineering.

    Why it matters: Anthropic’s checkout volume and repository size offer concrete context for understanding how increased coding activity can put pressure on the infrastructure supporting code review and CI.

  36. TechCrunch · AI58

    Industrial Next wants factory robots to switch tasks in minutes, not weeks

    Industrial Next presented AI-driven robotic stations at TechCrunch Disrupt 2026 that it says let manufacturers retool production lines in minutes rather than weeks without traditional programming. Its proprietary foundation models connect vision, task context, and motion, with new tasks taught through simulation and demonstration. The startup announced a partnership with BMW Group Technology Office to evaluate dexterous manipulation for manufacturing, and co-founder Lukas Pankau said some stations already perform delicate assembly tasks for an undisclosed smartphone customer.

  40. Latent Space73

    Dyna Robotics’ Jason Ma Explains Dyna-2 and Hour-Long Physical Agent Workflows

    Dyna Robotics co-founder and Chief Scientist Jason Ma explains how Dyna-2 combines future video prediction with robot actions and how agentic harnesses support hour-long workflows. He says the model was trained on over one million hours of human video, with some tasks requiring as little as 45 minutes of demonstrations. The interview covers napkin-folding deployments at Din Tai Fung, on-device inference, and hardware durability and missing sensory data as constraints on broader robot capabilities.