NAll news

All news

Tuesday792 items

  1. China.com Technology
    AI score27

    2026 AI Design Tool Recommendations: Lovart 2.0 Advances the Complete Design Workflow with Design Agent Arti

    Lovart 2.0 launched its personal design Agent Arti on October 19, 2026, enabling AI to participate in early-stage design research, canvas inspection, and feedback-based revisions, rather than merely generating images from prompts. Arti can remember creators' preferences, gather references from dozens of sites such as Google, Pinterest, and Behance to generate research reports, and distill repetitive workflows into reusable Skills. The article also lists Jimeng AI as a tool suited for quickly producing creative images and short video materials.

    Same story · 10 reports

    #Product updates#Agents

  2. China.com Technology
    AI score28

    How Lovart 2.0 Supports Custom Text Editing and Layout Adjustments for AI Posters

    Lovart 2.0 launched its personal design Agent Arti on October 19, 2026, linking Reference, Design Research, a visual canvas, and Design Review to form a revision workflow that takes a poster from first draft to a publishable version. Reference supports images, videos, PDFs, Word documents, Markdown, and web links, making it easy to centralize brand assets and copy references; Design Review lets collaborators comment on specific design elements via a share link. The article emphasizes that its adjustable layout is not equivalent to the character-by-character, layer-by-layer control of professional DTP software, and that strict print scenarios still require professional design software.

    Same story · 10 reports

    #Product updates#Agents#Image generation

  3. China.com Technology
    AI score22

    Building brand visual design from scratch: how Lovart 2.0's Arti design Agent connects the complete workflow

    Lovart 2.0, through its personal design Agent Arti, connects Design Research, Reference, a visual canvas, Design Review, and Skills into a complete brand visual design workflow. Among these, Design Research can generate research reports with sources, Reference supports centralized management of materials such as images, videos, PDFs, Word documents, Markdown, and web links, and Design Review provides a comment feedback process and can distill repeated methods into reusable Skills. Before formal commercial use, authorization for logos, fonts, likenesses, and third-party materials still needs to be verified separately.

    #Tutorials / practice#Image generation#Agents

  4. China.com Technology
    AI score9

    A Must-Learn Lesson for Brands in the GEO Era: How Soft-Article Publishing Network Gets Brand Information Adopted by AI

    Soft-Article Publishing Network uses a media matrix of 100,000+ outlets, four-step self-service publishing, and an editor-in-chief-level writing team to help brands place information in sources AI is more willing to trust. The platform emphasizes authoritative sources and consistent messaging, supports 24-hour self-service publishing, same-day output for standard articles, and provides performance review plus one-on-one dedicated follow-up.

    #Industry news

  5. China.com Technology
    AI score10

    From Indexing to Citation: Why Brand Press Release Placement Must Shift to GEO in the AI Era

    The acceptance criteria for brand press release placement are shifting from "being indexed by search engines" to "being cited and trusted by AI Q&A tools such as Doubao and DeepSeek." Articles that are merely indexed are unlikely to appear in AI answers. AI prefers authoritative news media, vertical industry sites, and continuously updated content platforms. Toumeiwang, Ruanwen Fagao Wang, and Youmeihui have already built out related services around GEO (Generative Engine Optimization). Brands can optimize their source structure through simulated testing and auditing existing sources; those who lay the groundwork early are more likely to be recognized and remembered by AI.

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

  6. China.com Technology
    AI score28

    Lenovo reshapes how individuals and enterprises work with AI-native devices and enterprise AI agents

    Lenovo Group is advancing its AI-native transformation through Tianxi, AI Host, and LeXiang, with LeXiang's sales conversion rate in Lenovo China's fiscal year 2025/26 up 23% compared with the traditional official website. LeXiang 4.0, launched on June 30, is supported by the Phoenix agent runtime layer and can call product, order, and service systems to complete transaction and after-sales tasks. Lenovo China has also organized its software R&D teams into collaborative teams of "full-stack engineers + five agents."

    #Industry news#Agents#OpenAI

  7. East MoneyOpinion
    AI score53

    Xinhua Finance Investigation: The Commercialization Test After China's Large Model API Calls Rank First Globally

    A Xinhua Finance investigation points out that, according to OpenRouter statistics, during the week of October 19 to 25, 2026, the weekly API call volume of Chinese AI large models reached 93.54 trillion tokens, ranking first globally for 26 consecutive weeks, but whether call volume can be converted into sustainable revenue remains a real-world test. The report says that free traffic-generation and test calls account for a not-insignificant share, and that after deducting computing power, labor, and operations and maintenance costs, some API orders yield meager profits or even run at a loss, while the industry has yet to form a unified token pricing standard. Alibaba's AI cloud and computing power service revenue for the quarter ended June 30, 2026, was 48.437 billion yuan, with AI-related product revenue of 12.376 billion yuan, while a large number of small and medium-sized model vendors remain in the pilot stage.

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

  8. East Money
    AI score20

    Chen Sijing of Guotai Haitong Asset Management: The tech sector is undergoing a "20/80 shift," with excess opportunities moving beyond computing power

    Chen Sijing, Deputy General Manager of the Equity Investment Department at Guotai Haitong Asset Management, believes that the tech sector is undergoing a "20/80 shift": computing power accounts for only 10%–20% of the stocks in the tech sector, and excess returns are hidden in the other 80% beyond computing power. He is more optimistic about the long-term value of on-device AI, AI software, and semiconductors, and adopts a Tick-Tock strategy modeled on Intel's cadence, switching between "generating returns" and "controlling drawdowns." Guotai Haitong Growth Select, for which he is the proposed fund manager, is currently being offered.

    #Industry news

  9. East MoneyOpinion
    AI score56

    Morgan Stanley CEO Pick Expects AI Capital Expenditure to Reach $10 Trillion to $15 Trillion Over the Next Decade

    Morgan Stanley CEO Ted Pick expects that AI-related capital expenditure over the next decade could reach $10 trillion to $15 trillion, making it the most influential economic force since the birth of the internet. He estimates that hyperscale cloud service providers may invest $800 billion to $1 trillion annually over the next 7 to 12 years, and says current progress is equivalent to the second inning of a nine-inning game, with $2 trillion to $3 trillion in capital already committed or actually deployed. He also estimates that the productivity dividend brought by AI could raise the growth rate of U.S. real GDP by 50 to 70 basis points per year.

    Same story · 4 reports

    #English text is being prepared.#Industry news

  10. East Money
    AI score80

    DeepSeek V5 plans to pretrain on Ascend chips

    According to sources in the industry chain, DeepSeek's next-generation large model, DeepSeek V5, plans to conduct pretraining based on Huawei Ascend chips; previously, Ascend was mainly used for inference tasks of DeepSeek models. The report mentions that after the release of DeepSeek-V4 in April 2026, Huawei stated that the full range of Ascend supernode products could fully support the V4 series models; on September 30, DeepSeek open-sourced infrastructure components for the Ascend computing platform, covering TileLang compilation tools, DeepGEMM, FlashMLA, DeepEP, and others. Currently, the implementation of domestic computing power remains concentrated on the inference side; the full-parameter post-training of DeepSeek-V4-Pro previously completed on the Ascend 910C cluster did not involve foundational pretraining.

    Same story · 3 reports

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

  11. QbitAI
    AI score22

    AI data platform OceanBase wins World Internet Conference Outstanding Contribution Award

    OceanBase, incubated by Ant Group, received the "World Internet Conference Outstanding Contribution Award" at the 2026 World Internet Conference. The conference's award citation stated that it has built AI data infrastructure and served the development of the global digital economy. In 2026, OceanBase released a lake-and-database integrated AI database and built an AI data platform system covering AI databases, AI lake-and-database, and DataPilot, with related capabilities already implemented in scenarios of dozens of customers including Ant Afu, Lingguang, and Li Auto. Calculated by 2025 revenue, OceanBase ranks first in the distributed database market in Asia and Oceania and has served more than 4,000 customers worldwide.

    Same story · 3 reports

    #Industry news#English text is being prepared.

  12. QbitAI
    AI score27

    Apple reportedly developing J450 privacy camera; China Telecom TeleAI says it launched a similar product 8 months ago

    Bloomberg reporter Mark Gurman reports that Apple is developing a smart home camera codenamed J450 that does not record video, instead analyzing the environment via AI and pushing text descriptions to users to protect privacy. The China Telecom Artificial Intelligence Research Institute (TeleAI) says that its TeleAI privacy camera, built on AI Flow, was launched as early as February this year for the public safety sector, capable of real-time risk behavior alerts while protecting privacy.

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

  13. Phoenix Tech
    AI score42

    OpenAI Hires Former U.S. Army Secretary Driscoll as Senior Advisor for National Security Modernization

    OpenAI has hired former U.S. Army Secretary Dan Driscoll as its newly created senior advisor for national security modernization, focusing on the application of AI in national security and cybersecurity. He will work with OpenAI management and external partners to promote the adoption of related technologies by U.S. allies and participate in work related to "democratic AI"; previously, OpenAI had advanced cyber defense and critical infrastructure protection through the Daybreak project.

    Same story · 7 reports

    #Industry news#OpenAI

  14. Phoenix Technology
    AI score71

    Report: DeepSeek V5 Plans to Use Huawei Ascend Chips for Pretraining

    According to a report by the STAR Market Daily, multiple industry chain sources say that DeepSeek's next-generation large model, DeepSeek V5, plans to conduct pretraining based on Huawei Ascend chips. Previously, Ascend chips were mainly used for inference tasks of DeepSeek large models.

    Same story · 3 reports

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

  15. Phoenix TechnologyProduct release
    AI score58

    OpenAI plans to merge ChatGPT chat and work modes, Codex to remain independent

    OpenAI plans to merge the Chat and Work tabs in ChatGPT, integrating everyday conversation and complex task handling into a single interface, while Codex, which is oriented toward programming tasks, will retain an independent entry point. After the adjustment, users will be able to move from Q&A to multi-step tasks without switching modes, and within the same conversation call files and plugins, view task progress, and intervene; OpenAI plans to first open testing to some users and then gradually expand the scope. Usage rules will also be adjusted accordingly: starting in December, ChatGPT and Codex usage for Plus, Pro, and Business users will be calculated together and uniformly measured in tokens, while ordinary instant chat will not be counted.

    Same story · 13 reports

    #Product updates#OpenAI#Agents

  16. Testing Catalog
    AI score25

    November 3 AI Daily Brief: Anthropic Q3 Revenue $17 Billion

    Anthropic's revised IPO filing shows Q3 revenue of $17 billion and compute commitments of $895 billion, and the Claude Messages API message threads have entered public beta. OpenAI released LifeVal, a benchmark for everyday assistant tasks, and raised its bug bounty to as much as $172,000; Perplexity open-sourced pplx-embed-v2-context, a vision-language embedding model in 9B and 0.6B sizes. Tencent launched its Hunyuan research agent Hyra 2.0, and Alibaba's Qwen Go opened a free tier, providing each user with a dedicated VM to run agent tasks.

    Quoted post🚨 AI News | TestingCatalog · @testingcatalog

    DAILY AI BRIEF 🗞 — Nov 2 GOOGLE 🔥: - Gemini 4 Carbon is reportedly being tested on Arena as Argon 4 (via Waguri_Kaoruko8). - Antigravity may get cloud execution support next week. - Antigravity's "Chief of Stuff" agent will get Gemini Live support. - Antigravity is getting a new security notice about coding agent risks. - Gemini Notebook may add a "Comic" artifact type. OPENAI 🔥: - GPT-6.2 Astra is showing up in Codex and ChatGPT Work usage analytics (via StatsWire). ANTHROPIC 🔥: - A hidden Claude Mythos 5.5 mention was spotted on Anthropic's website. SPACEXSI 🔥: - SpaceXSI will enable text watermarking on Nov 7. DEEPSEEK 🔥: - DeepSeek open-sourced dsh-superagent, a research prototype of an autonomous coding agent. - DeepSeek V4.1 Pro is showing up in benchmark setups (via LuminaBench). MEITUAN 🔥: - Meituan released LongCat-2.5, a native multimodal model for agents, API only. SEMIANALYSIS 🔥: - SemiAnalysis released GDPx, an agentic benchmark where the top model scores 12%. HERMES 🔥: - The Hermes mobile app opened beta testing on iOS and Android, and both betas are now full. * Used Grok to compose this brief, cherry-picking the news and doing some post-editing.

    #Industry news#OpenAI#Anthropic

  17. cnBeta
    AI score55

    UK NCSC Warns AI Will Amplify Cyber Threats to Legacy Systems

    The NCSC, part of the UK's Government Communications Headquarters (GCHQ), has released a report concluding that it is highly likely hostile actors will adopt artificial intelligence technology, enabling them to carry out large-scale ransomware attacks and targeted social engineering fraud at lower cost in the future and to quickly discover and exploit vulnerabilities. NCSC Chief Executive Richard Horne said new technologies are amplifying fundamental security weaknesses that organizations have long neglected. The report notes that some models from Google, Anthropic and OpenAI have already demonstrated the ability to find software vulnerabilities and analyze code flaws. The UK government also announced investment plans, including 60 million pounds for the first phase of the National Cyber Shield project, 70 million pounds to establish the National Exposure and Real-time Vulnerability Centre (NERV), and the creation of an AI cybersecurity centre of excellence in Manchester operated by the NCSC.

    #Safety / alignment#Industry news#OpenAI

  18. Hugging FaceFirst-party
    AI score51

    Hugging Face proposes the Transformers Model Integration Contract (MIC)

    Hugging Face proposes the Transformers Model Integration Contract (MIC), which packages model tests into three parts—a scaled-down fixture model, runnable Markdown usage examples, and manually reviewed baseline outputs—and runs fast integration tests on CPU. The team used the frozen MIC to replay five months of Transformers commits; the UMT5 fixture consistently reported a log-probability deviation of 1.22 and a training loss deviation of about 0.20 from when the regression was introduced on July 30, 2026 until it was fixed on September 28. The scaled-down report for Llama 4 Scout shows that 10.86 billion parameters were reduced to 44.7 million, the vision stack was reduced from 34 layers to 2 layers, and the language stack was reduced from 48 layers to 4 layers. As of November 1, 2026, this CI job only reports failures and does not block PRs.

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

  19. Shanghai Observer
    AI score22

    Shanghai Open University and Shanghai Education Television Launch "AI Literacy 100 Points" 100-Episode Short Video Course for Seniors

    The 100-episode short video course "AI Literacy 100 Points" (Senior Edition), jointly created by Shanghai Open University and Shanghai Education Television, premiered on November 2. Aimed at people aged 55 to 75, each episode uses five minutes to explain an AI application in a real-life scenario. The course is organized into seven major sections: basic understanding, everyday tools, life scenarios, health and wellness, family and relationships, fraud prevention and safety, and the intelligent connection of all things. Fraud prevention and safety is a standalone section, teaching about new types of scams such as AI face-swapping, AI voice-changing, and fake customer service. The course airs on Shanghai Education Television's large screen with two episodes per day from Monday to Friday, and is simultaneously released on the official website, WeChat public account, video channel, Douyin account, and the "One Network for Learning" platform.

    #Tutorials / practice#Video

  20. TechCrunch AIProduct release
    AI score61

    Apple Watch audio intelligence features enter developer testing, including Live Rewind and Siri Recaps

    Apple's Audio Intelligence features for Apple Watch have entered developer testing with iOS 27.3, including four optional features: Sound Recognition, Music Recognition, Live Rewind, and Siri Recaps. Live Rewind lets users double-tap the Digital Crown to review the text of the last 15 seconds of conversation, with unsaved clips automatically deleted after 30 seconds; Siri Recaps only generates key-point summaries, neither saving raw audio nor performing verbatim transcription, with unsaved summaries deleted within 7 days. Audio is processed in the watch's Secure Enclave, converted into condensed text via iPhone, and then sent to Private Cloud Compute to generate summaries. It is available for Apple Watch Series 12 and Apple Watch Ultra 4.

    Same story · 5 reports

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