NAll news

Friday1456 items

  1. Alibaba Cloud Developer Community AI
    AI score34

    How EMR Serverless Spark Supports Ultra-High-Concurrency Python Distributed Tasks

    Alibaba Cloud EMR Serverless Spark supports thousands of Python task submissions per second and the concurrent execution of hundreds of Ray Actors through three layers of mechanisms: declarative resource orchestration, isolated execution contexts, and observability operations contracts. Its Fusion engine has a built-in Arrow zero-copy transport channel and, together with Celeborn Shuffle Service, achieves effective data throughput of 800+ MB/s over gigabit networks. The platform supports the coexistence of multiple Python versions from 3.9 to 3.12; enabling spark.sql.pyspark.udf.simplified=true can reduce UDF serialization overhead by 40%.

    #Tutorials / practice#English text is being prepared.#English text is being prepared.

  2. 36Kr · AIProduct release
    AI score71

    Neuralink Connects Grok to the Brain: ALS Patient Controls AI Agent with Thoughts

    Neuralink has announced its latest progress: Kenneth, a patient with ALS, silently thinks "Hey Neuralink" and formulates a task in his mind, which then wakes up the Grok agent on his computer to autonomously execute the operation. The team conducted a 72-hour development sprint at the subject's home, producing 8 functional prototypes, including neural voice activity detection (VAD) that requires no manual switch, a system-level brain-computer input method called BCI Keyboard Dictation, and a mobile app called Bloom. Kenneth also used speech generated by the brain-computer interface to connect to the iPhone call audio stream, making a phone call to his wife with his thoughts for the first time in seven years.

    Same story · 3 reports

    #Product updates#Agents#Embodied AI

  3. 36Kr · AI
    AI score68

    Google's stock price plunged over 11% after its earnings report, as AI investment devours cash flow and buybacks drop to zero

    After Alphabet released its earnings report, its stock price plunged more than 11% in a single day, with about $500 billion in market value wiped out; operating profit was $40.6 billion, below the market expectation of $43.2 billion. AI investment continues to devour cash flow, and stock buybacks fell to zero in the first three quarters of 2026, with the company turning to debt issuance and share sales to raise funds; search query growth hit a five-year high, but search revenue growth slowed to 12%, and the lower monetization rate of AI Mode has become a focal point of market concern. Cloud business revenue doubled year over year to $30.7 billion, and its profit margin rose to 36.2%, but the loss from shared AI research and development widened from $4.5 billion to $7.7 billion.

    #Industry news#Google#Microsoft

  4. 36Kr · AI
    AI score32

    Torani, 101 Years Old and Never Having Laid Off Employees: Uses AI but Does Not Lay Off Older Employees, Revenue Expected to Exceed $800 Million in 2026

    The American syrup manufacturer Torani has never laid off employees in its 101-year history. CEO Melanie Dulbecco said the company also uses AI and automation, but only reduces new hiring and does not cut existing staff. During the pandemic, the company's revenue once fell 20%; by cutting marketing expenses and shifting to online retail, full-year revenue in 2020 instead grew 15%. Torani currently has more than 500 employees, and over the past 35 years its annual revenue has grown by an average of about 20%. It is expected to exceed $800 million in revenue in 2026.

    #Industry news

  5. Jiwei.com
    AI score18

    Jingcun Technology Supports On-Device AI Applications with LPDDR5X Memory Solutions

    Jingcun Technology is competing for the Best Solution of the Year Award and the Outstanding Innovation of the Year Award at the 8th IC Fengyun List with its LPDDR5X high-bandwidth, low-power memory solution for on-device and edge-side AI. The solution supports up to 9600Mbps; the LPDDR5/5X series covers 1GB–32GB capacities, DQ16/32/64 bit widths, and FBGA245/315/496 and other packages; at the same bit width and channel count, 9600Mbps offers a 50% increase in theoretical data transfer rate over 6400Mbps.

    #Industry news#English text is being prepared.

  6. Jiwei.com
    AI score31

    Chixin Semiconductor Breaks Through with Four Major UWB Chips, Achieving Import Substitution with 22nm Full-Stack Architecture

    Leveraging its original "positioning + communication + radar" trinity architecture and advanced 22nm process, Chixin Semiconductor has launched four UWB SoC chips—CX100, CX320, CX520, and CX200—covering consumer electronics, IoT, and automotive front-installation markets. Among them, the automotive-grade CX520 is based on TSMC's 28nm process and is the first domestic UWB chip to pass the full FiRa Core 3.0 certification suite, with a success rate of over 99.9% in resisting relay attacks. The CX100 and CX320 have already shipped at scale, the CX520 has entered the supply chains of leading domestic and international automakers, and the CX200 has entered the pre-mass-production stage.

    #Industry news#English text is being prepared.

  7. Jiwei.com
    AI score31

    OmniVision Group's Q3 2026 revenue reached 8.124 billion yuan, a record high, with net profit up more than 50% quarter-on-quarter

    OmniVision Group's Q3 2026 revenue was 8.124 billion yuan, up 3.8% year-on-year, setting a new single-quarter record; net profit attributable to shareholders was 1.079 billion yuan, up 50.53% quarter-on-quarter, and gross margin recovered to 30%. Cumulative revenue for the first three quarters was 22.149 billion yuan, with semiconductor design business revenue of 16.346 billion yuan, accounting for 73.80%, while the share of intelligent driving, professional imaging, machine vision and robotics product lines continued to rise. The company expects Q4 revenue of 7.759 billion to 8.324 billion yuan, up 9.71% to 17.71% year-on-year.

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

  8. Jiwei.comProduct release
    AI score65

    Iluvatar CoreX's Tiangai 300 Benchmarked Against H200 in Real Tests, and On-Device Terminal Tongyang TY1200 Released

    Iluvatar CoreX held its Shanghai media open day on October 28, where it opened the Tiangai 300 for live demonstrations for the first time, saying that in third-party real tests its single-card end-to-end throughput reached 1.05 times that of the H200. The company also provided lab data such as 64K-context Attention at 1.05 times that of the H200 and MoE throughput under GLM-5.3 at 1.15 times, and showcased its cloud-edge-device layout, with the on-device terminal Tongyang TY1200 rated at 300 TOPS INT8 dense computing power.

    Same story · 4 reports

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

  9. heise online
    AI score33

    Auslegungssache 171: EU Digital Omnibus accused of favoring tech companies, vzbv discusses…

    The EU Digital Omnibus is intended to amend the GDPR and the E-Privacy Directive in one go. Michaela Schröder of the German consumer organization vzbv criticized the proposal in episode 171 of the c't podcast Auslegungssache, saying it is packaged as "clarification" but actually contains major changes. She pointed out that the Council position agreed by member states on October 21 removed the machine-readable cookie preference signal and eliminated the right to object unconditionally to the use of personal data for AI training, so cookie banners may therefore continue to exist.

    #English text is being prepared.

  10. heise online
    AI score43

    Alphabet's market value evaporated by about $500 billion in a single day, and Amazon's stock price came under pressure due to AI investment

    Alphabet's stock price fell about 12.5% in a single day, with its market value shrinking by about $500 billion, the largest single-day drop in the company's history, due to rising AI spending and plans to further increase AI data centers this quarter. Amazon's revenue and profit continued to grow, and AWS grew significantly, but AI infrastructure investment pushed free cash flow down to a negative range of tens of billions of dollars, and its stock price edged lower.

    #Industry news#Google#Microsoft

  11. heise online
    AI score50

    Reddit will require users to log in to browse content

    Reddit CEO Steve Huffman announced that the platform will change its nearly two-decade-old practice of allowing content to be read without an account, first testing the login requirement in markets outside the US. In the third quarter, Reddit had 130.3 million daily active users, of whom 76.8 million were not logged in; Huffman said most unauthorized scraping comes from logged-out visits, and stressed that registration does not require revealing one's real identity. Reddit previously announced that it will discontinue RSS on November 13, shut down its public API by March 2027, and has sued Anthropic and Perplexity for scraping content without permission.

    Same story · 4 reports

    #Industry news#English text is being prepared.

  12. heise online
    AI score24

    Low-cost shared AI account subscriptions: heise tests GamsGo

    heise tested shared Pro accounts for Claude, Perplexity and Suno purchased through GamsGo; all three accounts worked normally, but sharing the same account with heavy users can lead to noticeable restrictions. GamsGo offers subscriptions to various AI services at prices far below the regular rate, but the original providers' terms and conditions (AGB) prohibit such shared use; the article also cites lawyers' opinions to analyze the legal risks users face, mainly at the level of data protection law.

    #English text is being prepared.#OpenAI#Anthropic

  13. heise online
    AI score36

    Anthropic Becomes SpaceX's Largest Customer, Replacing the U.S. Government

    According to Bloomberg, Anthropic has become SpaceX's largest single customer, corresponding to "Customer B" in SpaceX's latest financial report and contributing about one-third of revenue. Anthropic reached a multi-billion-dollar partnership with SpaceX this spring, obtaining most of the AI compute of xAI, which was merged into the latter, and SpaceX thereby achieved billion-level profit for the first time after its record-breaking IPO. SpaceX is shifting its focus to AI, Crew Dragon is only planned to carry crew until 2030, and Falcon 9 may be retired in 2028.

    #Industry news#English text is being prepared.

  14. TMTPost · AGIOpinion
    AI score53

    Who Is Really Eating the AI Dividend: Computing Power Investment and Profit Distribution as Seen from Silicon Valley Financial Reports

    TMTPost, after reviewing the financial reports of giants such as Meta, Amazon, and Google, pointed out that the AI dividend has not landed evenly across all links: Anthropic's revenue in the first half of 2026 was $16.286 billion, up 1,300% year on year, but computing power and infrastructure costs reached $8.945 billion, accounting for 55% of revenue; OpenAI's operating loss in 2025 was $20.92 billion. The article also cited a McKinsey survey saying that 80% of respondents believed AI improved efficiency, while only 37% believed it made a positive contribution to their company's EBIT, and mentioned that a Stanford study showed a 19% relative employment gap among young employees aged 22 to 25 in occupations highly exposed to AI.

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

  15. TMTPost · AGI
    AI score55

    Tencent Releases WorkBuddy Enterprise Edition and Buddy One, Proposing Four Hurdles for Enterprise Agent Deployment

    At Tencent's Global Digital Ecosystem Summit on October 29, Tencent released the observability center of WorkBuddy Enterprise Edition and Buddy One, a cross-application collaboration system for individuals; the latter uses WorkBuddy as its workbench to connect Tencent Docs, Tencent Meeting, ima, and Tencent Drive. Dowson Tong, Senior Executive Vice President of Tencent, said the biggest problem in enterprise Agent deployment is not that models are insufficiently powerful, but that there are too many breakpoints, and AI cannot connect to internal enterprise systems to get data. Wang Huixing, CTO of Tencent Cloud, summarized the deployment thresholds as four hurdles: it runs, it can be trusted, it can be measured, and it is easy to deploy, and said WorkBuddy Enterprise Edition has already covered half of China's Fortune 500 companies, with cumulative task requests exceeding 9 billion.

    Same story · 3 reports

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

  16. gihyo.jpProduct release
    AI score52

    Gemini Notebook Can Add Select U.S. and India Textbooks for Free

    On October 29, 2026, Google announced that Gemini Notebook (formerly NotebookLM) can add hundreds of textbooks for free, with sources including U.S. and Indian publishers such as OpenStax, Open Textbook Library, and S Chand Publishing. Google's Josh Woodward said on X that more than 700 textbooks are available for free. Users open the catalog by selecting Textbooks under Add sources within a notebook, can filter by subject, level, and country, and can ask questions based on textbook content, view citation links, and generate study guides, practice quizzes, and flashcards.

    Same story · 5 reports

    #Product updates#Google

  17. Unite.AI
    AI score55

    &Truth Raises $70 Million Seed Round to Build a General-Purpose Code Verifier with Lean

    &Truth emerged from stealth on October 29, disclosing a $70 million seed round and plans to build a general-purpose verifier capable of reading code in any programming language and mathematically proving that it is safe, correct, and reliable. The company is built on the open-source proof system Lean and the Mathlib library of more than two million lines, and says its proof automation technology can shorten proof generation from years or months to hours or minutes; it also plans to open-source its foundational library. Co-founders include Stanford University professor Clark Barrett and Ilya Sergey, founder of the VERSE Lab at the National University of Singapore, and the company is headquartered in Palo Alto and Singapore.

    Same story · 2 reports

    #Industry news#Safety / alignment#Open-source ecosystem

  18. English text is being prepared.Model releases
    AI score64

    AgiBot Releases GO-3 Preview Embodied Foundation Model, First to Propose the Worken Workload Metric

    AgiBot has released its new-generation embodied foundation model GO-3 Preview, which adopts its self-developed WRAM architecture to integrate task understanding, future state prediction, and action generation into a unified model, and is the first to propose the Worken evaluation framework for measuring a robot's actual work output. AgiBot says the model ranks first overall in three embodied intelligence benchmarks: RoboDojo, RoboTwin, and RoboColiseum. Worken equals the product of the number of active robots, autonomous working hours, and the output efficiency ratio; AgiBot also released the AHA3 hardware platform and Lingqu OS 1.0 at the same time.

    Same story · 12 reports

    #Model releases#Embodied AI#English text is being prepared.

  19. Intelligent Manufacturing Network
    AI score8

    The 23rd Zhengzhou Industrial Fair 2027 will be held from May 11 to 14 at the Central Plains International Convention and Exhibition Center

    The 23rd Zhengzhou Industrial Fair 2027 is scheduled to be held from May 11 to 14 at the Central Plains International Convention and Exhibition Center (Airport Economic Zone), with an exhibition area of 100,000 m² and ten major specialized exhibition zones, covering core tracks such as CNC machine tools, industrial automation, and robotics, and is expected to gather more than 120,000 professional buyers. During the same period, the main forum of the Central and Western China Intelligent Equipment Manufacturing Development Conference and several sub-forums will be held, and a special on-site machine purchase subsidy of up to 5,000 yuan will be launched.

    #Industry news

  20. Intelligent Manufacturing Network
    AI score15

    The 11th World Intelligent Manufacturing Expo 2026 Opens in Nanjing on November 19

    The 11th World Intelligent Manufacturing Expo 2026 will be held from November 19 to 21 at the Nanjing International Expo Center, under the theme "Intelligence Leads Change · AI Reshapes Manufacturing," with an exhibition area of 52,000 square meters and more than 400 companies participating. The expo features four major themed halls—Halls 4, 5, 6, and 7—and for the first time is finely divided by sub-sectors, covering intelligent robots, smart factories, international industry-education integration, and the intelligent manufacturing industry chain. Companies such as KUKA, Estun, Siasun, Nangang, XCMG, and Siemens will exhibit, and an outdoor plaza will host a live unmanned technology experience zone, showcasing L4 unmanned heavy trucks with automated loading and unloading and platooning.

    #Industry news#Embodied AI#Agents