NAll news

All news

Monday595 items

  1. Sina Tech
    AI score25

    Hassett Downplays AI Financing Concerns: No Large-Scale Sell-Off Expected

    Hassett downplayed concerns about AI financing, saying that while the market may see a minor pullback, he does not believe at all that a wholesale sell-off will occur. He noted that some companies are heavily indebted, but cash-rich giants hold related equity and debt, and would step in if a company runs into trouble or seizes the opportunity to take over—though he added that it may not come to that.

    #Industry news

  2. Sina Tech
    AI score56

    OpenAI pledges that its self-built U.S. data centers can cut power load during extreme weather

    OpenAI announced the launch of the "Grid Guardian Plan," under which data center campuses it directly builds in the United States will proactively reduce electricity consumption when extreme weather or other emergencies put stress on the grid. The plan extends to the entire country a commitment already made for a planned campus in Effingham County, Georgia; OpenAI pledges to cut up to 1 gigawatt of power load when the grid needs relief, roughly equivalent to the electricity consumption of 700,000 households. John McCarrick, OpenAI's head of energy policy, said this reflects a shift in industry thinking and technology, as well as adjustments made after incorporating community feedback.

    Same story · 2 reports

    #Industry news#OpenAI#English text is being prepared.

  3. cnBeta
    AI score58

    OpenAI launches Grid Protectors Program, data centers can cut up to 1 gigawatt of load in extreme weather

    OpenAI announced the "Grid Protectors Program," extending a commitment for a planned data center campus in Effingham County, Georgia, to the entire United States, with the ability to cut up to 1 gigawatt of electricity load when the grid needs relief, roughly equivalent to the electricity consumption of 700,000 households. OpenAI said it will negotiate with power companies to determine the scale of reducible load, response speed and duration, and details for each campus will be announced separately once finalized. John McCarrick, OpenAI's head of energy policy, said this reflects a shift in industry thinking and technology, and noted that a year ago some large cloud providers still insisted on the "five nines" reliability standard of 99.999% continuous uptime.

    Same story · 2 reports

    #Industry news#OpenAI#English text is being prepared.

  4. Shanghai Observer
    AI score23

    Shanghai Grandpas and Aunties Take the Stage at an International Film Festival with AI Videos, as AI Trend Courses in Senior Education Gain Popularity

    Two Shanghai senior learners with zero prior experience, Fang Zigen and Chen Dongxia, had videos they created with AI selected for the Hong Kong Bauhinia International Film Festival and the Hong Kong International Short Film Festival, winning awards including Best AI Short Film and Best Art Award. The Second Shanghai Senior Learning Course Exhibition released its first batch of 20 premium courses, about 40% of which are AI trend courses, and simultaneously launched the 100-episode course "AI Literacy 100 Points" aimed at the 55–75 age group. Data shows that 69.6% of surveyed seniors are willing to participate in AI learning.

    Same story · 3 reports

    #Industry news#Tutorials / practice#Video

  5. Shanghai Observer
    AI score29

    Siemens Healthineers showcases nearly 20 innovative products at the CIIE, with the world's first photon-counting CT-guided PCI solution making its debut

    Siemens Healthineers exhibited nearly 20 innovative products at the CIIE, with the world's first photon-counting CT-guided PCI solution making its first appearance, integrating photon-counting CT with the ARTIS genio floor native AI spectral imaging platform to achieve seamless connectivity from imaging diagnosis to interventional treatment. ARTIS genio floor completed its Asian premiere, and the Halcyon Evolve intelligent ring-based individualized radiotherapy platform had its global premiere, with fine beam conformal capability improved by 5 times. The domestically produced photon-counting CT NAEOTOM Alpha.Prime demonstrated high-precision radiotherapy simulation and positioning functions, with a supply chain localization rate exceeding 80%.

    #Industry news#Multimodal

  6. Shanghai Observer
    AI score10

    IMX International Music Season concluded at 1933 Old Millfun in Hongkou Music Valley, exploring the direction of traditional music in the AI era

    The IMX International Music Season 2026 and the "2026 Shanghai Cultural and Creative Competition Music Track" achievement event concluded at 1933 Old Millfun in Hongkou Music Valley. Over three days, more than 40 industry forums and roundtable discussions were held, with copyright institutions, music companies, and creators from more than ten countries and regions participating. In the competition's "Music Innovation+" track, 23 municipal-level finalist projects were awarded certificates of honor, and the solicitation phase received a total of 419 applications from around the world. The event also announced the "World Traditional Music Seed Bank Annual Outstanding Cases," with Migu Music, Zide Qin Society, and others selected.

    #Industry news#Multimodal

  7. Shanghai Observer
    AI score55

    AgiBot's 6 Genie G2 Robots Work 24-Hour Live-Streamed Shifts at a Convenience Store

    AgiBot launched a 24-hour livestream at an operating convenience store in Xuhui District, where 6 Genie G2 robots took on tasks including greeting customers, making coffee, serving prepared food, picking and restocking goods, and cleaning tables, working in 4 shifts under real-time supervision by the entire internet. As of 2:30 p.m. yesterday, the 6 robots had been on duty for 4.5 hours and completed a total of 337 tasks, including greeting customers 116 times, delivering and retrieving 84 items, and making coffee 30 times. AgiBot previously released the GO-3 foundation model and the embodied intelligence agent "Genie Master," and together with Tsinghua University and Infinigence AI proposed a collective experience learning paradigm, validated in a cluster of 96 robots, where the success rate rose from 86.7% to 98.3% as the number of robots participating in training increased from 16 to 96.

    Same story · 4 reports

    #Industry news#Embodied AI#Agents

  8. Shanghai Observer
    AI score33

    Shanghai Municipal Administration for Market Regulation Uses AI Agents to Safeguard the 9th CIIE: Food Safety Verified with One Photo, Elevators Understood with One Scan, Ads Identified with One Click

    The Shanghai Municipal Administration for Market Regulation has upgraded its comprehensive command platform for service and support for the 9th China International Import Expo (CIIE), embedding AI agents into the entire workflow of food safety, special equipment safety, and advertising inspections. The food safety vertical large model has been iterated into "Food Safety Smart Eye (One Photo, One Verification)", which is connected to 145 video feeds and conducts intelligent inspections of 55 catering units; high-risk units scoring below 80 are inspected on-site by regulators. All 424 elevators have achieved full coverage of QR code verification and remote monitoring, and AI advertising inspections have compressed full-coverage inspections from a whole day to about 7 hours.

    Same story · 2 reports

    #Product updates#Agents#Multimodal

  9. TechCrunch AI
    AI score45

    OpenAI will merge ChatGPT's Chat and Work tabs

    OpenAI announced that it will merge the Chat and Work tabs in ChatGPT, initially opening it to some users to gather feedback before rolling it out to everyone; after the merge, users can bring in files and plugins in one place, letting ChatGPT build websites, make presentations, or perform web operations. Codex will remain a separate experience in the ChatGPT desktop app, but ChatGPT and Codex will share the same token quota pool, and some users may more frequently have to wait for their quota to reset. Anthropic already merged Claude's chat and Cowork features into a single interface in September.

    Same story · 11 reports

    #Product updates#OpenAI

  10. Simon Willison
    AI score46

    The U.S. Census Bureau and "Super Intelligence": Wording Change in the BTOS Survey Questionnaire

    Due to Executive Order 14434, the U.S. Census Bureau's BTOS survey changed "Artificial Intelligence or AI" in its questionnaire to "Super Intelligence or SI" starting October 19, 2026. Previously, on November 17, 2025, the survey had already changed the scope of its questions from "producing goods or services" to "any business function," which raised the AI usage rate among firms with 250 or more employees from about 12% in September 2025 to about 40% in October 2026. The results of the first survey to use the SI wording are scheduled to be released on Thursday.

    #English text is being prepared.#English text is being prepared.#English text is being prepared.

  11. Tianyi CuiProduct release
    AI score55

    DeepSeek Harness Launches Creator Mode, Turning Ideas into Plugins

    DeepSeek Harness releases Creator Mode, allowing users to turn ideas into usable plugins, such as a design workspace, a sandbox game, or a 3D LLM visualization tool. Tianyi Cui reposted the news and asked for usage feedback.

    Quoted postDeepSeek Harness · @DeepSeekHarness

    Your Harness. You Define It. Meet Creator Mode in DeepSeek Harness. Turn ideas into plugins that you can use, from a design workspace to a sandbox game or a 3D LLM visualizer. What will you create? https://t.co/MlXHIyyA1o

    #English text is being prepared.#Product updates#Agents

  12. Artificial Analysis
    AI score40

    Artificial Analysis Adds GPU Rental Prices to the AA-AgentPerf Benchmark

    Artificial Analysis has partnered with Silicon Data to incorporate market-based GPU rental prices into the data center system performance benchmark AA-AgentPerf, adding a new "throughput per dollar per hour" metric and updating the rental price index for H100, H200, B200, and B300 on a daily basis. Among the systems for which rental prices are available, the NVIDIA B300 leads overall, while the B200 matches it on some models with faster output speed. The MI355X rental price shown in the chart is assumed by Artificial Analysis based on recent market quotes.

    #Product updates#Evaluations / benchmarks#English text is being prepared.

  13. CheDongXi
    AI score44

    After the democratization of intelligent driving, why is cockpit computing power hard to democratize: ModelBest's AI Box offers a pluggable upgrade solution

    The AI Box, built by ModelBest in partnership with Wutong Technology and Intel, is being delivered as an optional feature on the Changan Deepal L06. It comes with a built-in 4B VLM model and multiple on-device models, and took only three months from its debut in July to pre-installation mass production in October. Currently, high-end chips such as Qualcomm 8295 are concentrated in vehicles priced above 200,000 yuan, while vehicles priced around 100,000 yuan and below still mainly use Qualcomm 8155. The 3-4 month iteration cycle of cockpit large models is mismatched with the six-month to one-year mass production cycle of automotive SoCs. The density law proposed by researchers from Tsinghua University and ModelBest shows that the capability density of large models doubles roughly every 3.5 months, and industry competition is shifting from paper computing power to computing power utilization efficiency.

    #English text is being prepared.#English text is being prepared.#English text is being prepared.

  14. Shao Meng
    AI score62

    DeepSeek Search and Fetch End Free Tier, Billed at $0.75 per Thousand Calls

    The two web tool APIs, DeepSeek Search and Fetch, end their free tier starting today and switch to billing by the number of successful requests, at the same unit price of $0.75 per thousand calls. Billing only counts requests that return HTTP 200 and where Search returns non-empty results or Fetch actually retrieves page content; failures, empty searches, and timeouts are not billed, and the cost is independent of the number of returned items, tokens, and whether need_content is enabled. Rate limits are calculated per account: Search is 100 QPS and Fetch is 25 QPS, with Fetch also having a per-domain QPS cap; server-side tool calls initiated via DeepSeek Harness, Claude Code, and Codex are subject to the same rules.

    #Industry news#DeepSeek#English text is being prepared.

  15. Rohan Paul
    AI score40

    Sam Altman calls tokens a "fake unit"

    In a Cheeky Pint video, Sam Altman called tokens "a kind of fake unit": the number of tokens can be counted, but it cannot tell enterprises what this money bought. The author adds that the only defense of tokens comes from an information-theoretic perspective—model outputs are still mixed with noise; if the waste is removed, the token count would approach the Shannon limit, that is, the minimum amount needed to carry the actual information.

    #English text is being prepared.#Reasoning

  16. Rohan Paul
    AI score60

    TrendForce: NVIDIA Rubin Transition Begins, High-End GPU Lead Times Extended by 5 Weeks to 45 Weeks

    TrendForce data shows that with the start of the NVIDIA Rubin transition, high-end GPU lead times have been extended by 5 weeks to 45 weeks. Hyperscale cloud providers may get Rubin first, while enterprise buyers continue to use existing GPU platforms. On the memory side, arrivals of 96GB and 128GB RDIMMs have accelerated, while 32GB and 64GB modules remain severely tight; DRAM lead times have shortened by 1 week to 19 weeks, still more than twice normal levels. Because servers cannot ship until the slowest component arrives, Rubin's capacity ramp depends on substrates and capacitors whose lead times exceed balanced levels by 3x.

    #Industry news#English text is being prepared.#English text is being prepared.

  17. ModelBest OpenBMB
    AI score24

    ModelBest Launches MiniCPM Spark Program Season 2

    ModelBest OpenBMB has launched the MiniCPM Spark Program S2, calling for runnable projects from developers and researchers, with two tracks: the Open Build Track and the On-Device Agent Challenge. The designated models for this season are MiniCPM-OCR 1.0 / Jev and MiniCPM-V4.7, and another new MiniCPM model will be announced soon. Outstanding projects in the On-Device Agent Challenge can win a Mac mini M6; the first phase runs from November 2 to December 13, 2026.

    #Industry news#English text is being prepared.#Agents

  18. Phoenix TechnologyProduct release
    AI score53

    Tencent Hunyuan Releases AI Research System Hyra 2.0

    Tencent Hunyuan released the AI research system Hyra 2.0 on November 2, shifting its R&D focus from directly using AI to tackle scientific problems toward having general-purpose Agents participate in the development and upgrading of scientific models, research workflows, and software tools. The system splits research work into three layers—Overall, Mission, and Worker—and designs three separate storage mechanisms: project state, layered Agent memory, and a shared knowledge base called Books. During project retrospectives, transferable methods can be organized into skills for use in subsequent tasks. Previously, Hyra 1.0 drove open-source models to solve an open problem in additive combinatorics that had remained unresolved for over fifty years.

    Same story · 2 reports

    #Product updates#Agents

  19. Cailian Press
    AI score57

    Communities Across the U.S. Oppose Data Center Construction, Adding Uncertainty to Midterm Elections

    Twenty U.S. communities will vote on pausing or banning local data center construction, 13 of which are located in Republican-dominated districts. A POLITICO poll shows that 36% of Republican voters and 65% of Democratic voters oppose building data centers within 3 miles of their homes. Oracle's $165 billion Jupiter project in New Mexico faces delays, and an executive order in New York State pausing environmental permit approvals for large facilities has already delayed nearly 1GW of data center construction.

    #Industry news#English text is being prepared.#OpenAI

  20. Sina Tech
    AI score17

    Jiushi CEO Kong Qi Elected Vice Chair of the World Internet Conference Artificial Intelligence Professional Committee

    Jiushi CEO Kong Qi was elected vice chair of the professional committee at the 2026 plenary meeting and third chairpersons' meeting of the World Internet Conference Artificial Intelligence Professional Committee. The meeting was chaired by Wang Jian, chief chair of the professional committee and academician of the Chinese Academy of Engineering. Jiushi has long been deeply engaged in L4-level autonomous driving. In September 2026, it completed a "city-level physical AI" strategic upgrade and built an intelligent technology system centered on "Jiushi Brain AI Driver." To date, Jiushi's global unmanned fleet has exceeded 30,000 vehicles, with cumulative real operating mileage of nearly 300 million kilometers, covering more than 360 cities worldwide.

    #Industry news#Embodied AI