Skip to content
Live
Nov 8Sun
  1. Hugging Face Daily Papers24

    From Autonomous Agents to Cognitive Symbionts: Towards AI-Enhanced Human Intelligence

    Physical AI should enhance human intelligence through a lifelong cognitive symbiont, which compensates for cognitive gaps through perception, situational retention, and proactive assistance, requiring embedded governance. Large foundation models and personal agents can already complete multi-step tasks independently, but humans will still forget knowledge, ignore information, and lack historical references, as the machines that can provide help are outside the context.

  2. The Decoder75

    Vals AI benchmark shows the gap in network attack capabilities between open-source and closed-source models continues to narrow

    Vals AI's CyberOffense benchmark allows AI agents to autonomously attack simulated enterprise networks on isolated Kali Linux machines, GPT-6 Sol Daybreak in OpenAI restricted projects completes 80.8% of the target, open-source weighted model Deepseek V4.1 Flash 1017 reaches 72.8%, with a single run cost of about 3 dollars instead of about 22 dollars

Nov 7Sat
  2. Hacker News · 300+0

    Every Mac comes with a built-in (not publicly disclosed) family classifier

    The knowledge constructiond daemon in macOS 26.6.2 analyzes contact, calendar, photo, message, email, location, screen time, and other data on the device to build a knowledge graph containing 181 entity classes and 278 predicates, used to infer user and their family relationships. The graph is stored as an SQLite database in the ~/Library/IntelligencePlatform directory protected by TCC full-disk access.

  3. ITHome · AI35

    Tesla CEO Musk expresses affection for China again, praising the high level of the Chinese automotive industry

    During an interview, Musk stated his love for China, highly praising the competitiveness of the Chinese automotive industry, and said the Tesla Model Y produced in Shanghai is the best-selling electric vehicle in China. He specifically added that Chinese automotive companies are very impressive and believes the overall performance of Chinese AI large models is excellent, with top-tier unit computing power output performance globally.

  4. ITHome · AI65

    GitHub introduces a new Git storage architecture Rubicon: supports 5000 intelligent agents collaborating concurrently, processing 7200 pushes per second

    GitHub releases a new Git storage architecture Rubicon, internal testing supports a single code repository to accommodate 5000 intelligent agents for concurrent collaboration, processing 7200 pushes per second, and has now launched a small-scale targeted internal test, planning to gradually expand the scope of opening in early next year.

  5. ITHome · AI72

    Anthropic must continue to defend against the Claude Max plan usage misleading lawsuit, as the judge did not adopt its arguments

    U.S. federal magistrate judge Xie Wenhui ruled that Anthropic must continue to defend against the main claim that the Claude Max plan usage is misleading to consumers, and dismissed several other claims in the case. The plaintiff claims that the $100 and $200 per month Max plans, when calculated on a per-week basis, actually provide only 3.5 times and 6 times the usage of the Pro plan, respectively, rather than the advertised 5 times and 20 times.

    Why it matters: The ruling reveals the discrepancy between the promotional subscription limits and the per-cycle usage, allowing readers to understand the promotional boundaries of subscription-based AI products.

  7. ITHome · AI65

    Rajesh Nambiar, president of the National Association of Software and Services Companies in India, warned: AI may make it harder for fresh graduates to enter the tech industry.

    Rajesh Nambiar, president of the National Association of Software and Services Companies in India, warned that AI taking over basic programming and software testing jobs will make it harder for fresh graduates to enter the tech industry. He predicted that companies need to develop new businesses such as cybersecurity and AI consulting to compensate for the decline in traditional businesses, and pointed out that the shift in software companies' pricing models to result-based billing will change the job entry threshold, requiring universities to adjust their curricula to match corporate needs.

  8. ITHome · AI65

    Nvidia has provided billions of dollars in support for TSMC to expand its CoWoS advanced packaging.

    NVIDIA is providing billions of dollars in support for TSMC's expansion of CoWoS packaging capacity under its subsidiary TSMC Advanced Packaging, for the expansion of new clean room facilities and equipment procurement. TSMC is forcing chip design companies to seek external support due to capacity constraints, and CoWoS technology has been widely adopted by NVIDIA, AMD, Broadcom, and cloud giants.

  10. The Verge · AI80

    The METR report stated that Anthropic's models are self-accelerating their R&D, causing researchers' concerns.

    AI safety research organization METR released a report after an internal investigation of 11 weeks with Anthropic, stating that the capability improvement rate of Anthropic's models is about twice as fast as a year ago, achieving about two years of historical progress in the past year, and researchers are confident that the acceleration mainly comes from AI rather than computational power expansion.

    Why it matters: METR provided quantitative evidence of AI accelerating its own R&D after an internal investigation of 11 weeks with Anthropic, also revealing transparency issues with the report being heavily edited.

  11. The Decoder72

    Nous Research fine-tunes Qwen3.8-27B to launch Arete-27B, reducing the proportion of promotional answers from 22% to 0.9%

    Same story · 2 outlets

    Why it matters: The original text provides a comparison of the promotional rate and fact accuracy rate before and after fine-tuning, allowing readers to judge whether the political tendency of Chinese open-source models only stays on the surface.

  12. QbitAI65

    To get rid of the AI taste, humans are starting to write bad articles proactively

    Human writing ability shows reverse evolution, intentionally writing wrong words, illogical order, and missing components to fight the 'AI taste' generated by AI, leading to neglect of important language parts like rhetoric and punctuation. The latest models GPT-6 Astra and Gemini 3.1 Pro reduce the use of em dashes and mimic human style. AI detectors become language judges, affecting the fate of authors, and humans adapt to standards to pass detection, potentially leading to a hybrid human-machine language in the future.

  14. ITHome · AI35

    The Walker W900NB 30th Anniversary Limited Edition Over-Ear Bluetooth Headphones are now available: limited to 3000 sets, priced at 1159 yuan.

    The Walker W900NB 30th Anniversary Limited Edition Over-Ear Bluetooth Headphones are now available, with a limited quantity of 3000 sets priced at 1159 yuan. Equipped with 8 high-precision wide-angle silicon microphones, noise reduction depth up to -50dB, featuring a 6nm 6-core heterogeneous chip and Arm Ethos U55 independent NPU for end-to-end AI voice pickup and noise reduction, with a battery life of up to 110 hours.

  15. The Decoder78

    Nine AI entities attempt autonomous research on open problems at US national laboratories, none reaching expert standards

    A study made Claude Opus 4.8, Claude Opus 5, Claude Fable 5, Claude Opus 5.5, GPT-5.6 Sol, GPT-6 Astra.

    Why it matters: This benchmark uses 15 unpublished research questions to test 9 agents and provides common failure patterns in 135 submissions, which can be used as a reference for evaluating autonomous research capabilities.

  17. MarkTechPost60

    Buoyant releases Kinference: an open-source Kubernetes and vLLM inference control plane

    Buoyant has released Kinference, an open-source Kubernetes and vLLM inference control plane, with the current version being v0.1.0-alpha.7. It distributes requests in front of vLLM worker nodes through routing policies, token rate limiting, KV cache-aware load balancing, and tenant-specific metrics. In benchmark tests, for shared prefix workloads, the median TTFT decreased from 6.88 seconds to 0.31 seconds (95%).

    Same story · 2 outlets

    How the accounts differ:vLLM Blog mentions token limit, tenant usage and cost tracking, while MarkTechPost specifies token rate limiting and tenant metrics.

  18. ITHome · AI65

    Bill Gates on 'billionaire tax': Perhaps the most suitable tax reform in the AI era is to tax the rich

    Bill Gates believes that levying a tax on the rich similar to California's 'billionaire tax' is the most appropriate way in the AI era, because the founders of AI and related industries have seen a significant increase in wealth, and families holding shares in these companies can also expect substantial investment returns, exacerbating the wealth gap. Gates points out that one cannot rely on the rich to solve the wealth gap problem through charitable donations, and the government should implement progressive taxation based on wealth levels, as charitable donations are voluntary actions.

  19. ITHome · AI70

    Google's AI-PPG-ACC-HTN study published in Nature: Using consumer-grade smartwatches to detect hypertension without a cuff

    Google has released a multimodal AI system, AI-PPG-ACC-HTN, which does not require wristband calibration and passively detects persistent hypertension using a standard consumer-grade smartwatch. The results were published in Nature. The study included 196 participants across 5 clinical centers, using 24-hour ambulatory blood pressure monitoring as a reference. Initial screening showed a specificity of 90.0%, a positive predictive value of 80.6%, and a sensitivity of 65.8%. Diagnostic testing showed a positive predictive value of 96.7%, a specificity of 90.9%, and a sensitivity of 78.4%.

    Same story · 2 outlets
  20. The Decoder71

    Google AI detects hypertension using a regular smartwatch without a cuff

    Google Research's AI detects signs of hypertension without blood pressure cuff calibration, using only the pulse and motion sensors already in smartwatches, with related research published in Nature. The neural network learns pulse and heartbeat patterns by comparing short segments from nearly 500,000 hours of sensor data to identify which segments belong to the same person, then averages multi-day data to calculate the probability of hypertension.

    Why it matters: Readers can learn about a smartwatch blood pressure detection study that requires no cuff calibration, and its comparison results with traditional measurements in tests on 196 people.

  21. ITHome · AI65

    Zhang Yimou on AI short video output value reaching hundreds of billions and its challenge to the film industry

    Zhang Yimou states that the output value of AI short videos has reached hundreds of billions, posing a major challenge to the film industry, and is uncertain about how to respond when AI surpasses humans in the future. AI is reshaping the entire industry chain of micro-short dramas, from script creation to distribution and operation, with the authorities reiterating their support for real-person dramas and a regulated attitude towards AI dramas.

  23. ITHome · AI22

    China Popular Science Writers Association releases Top 10 popular science hot words for 2026, AI related occupy seven seats

    The China Science Popularization Writers Association released the top ten science popularization hot terms for 2026, with seven AI-related terms, including 'Artificial Intelligence +', 'Human-Machine Collaboration', 'Brain-Machine Interface', 'Intelligent Agent', 'Humanoid Robot', 'Token', 'Embodied Intelligence', and the other three hot terms are 'Science Toys', 'Technology Ethics', and 'Health Management'. The AI-related terms reflect the integration of technology into production and daily life and the public's need to understand cutting-edge technology. 'Science Toys' expand the form of science popularization, 'Human-Machine Collaboration' reflects AI's participation in science popularization creation, and 'Technology Ethics' emphasizes the responsibility of application.

  24. ITHome · AI67

    Apple reveals the humanoid robot system HALO: using the Qwen model to validate on the Yushu G1, achieving an average success rate of 91% for single tasks.

    Apple's research team reveals humanoid robot autonomous control system HALO, with an average success rate of 91% in 360 single task tests in 6 real environments; the system calls the Qianwen model for verification on the Yushu G1, achieving an average success rate of 70% in 10 tests of 3 combinations in a single long sequence compound task, and a success rate of 50% in the most complex task.

  25. Hugging Face Daily Papers65

    Drex open-source decision model released

    Drex open-source decision model returns the probability distribution of user options without autoregressive decoding, using supervised fine-tuning, RLAF, and consistency loss to improve accuracy, robustness, and latency. Its 35B parameter Max version achieves an industry-leading score of 67.52 on Decision Index 0.4.1, with a response time only 9% of GPT-6 Sol and 5% of Claude Opus 5.5.

  26. ITHome · AI65

    Google sued by parents in the US: Gemini accused of helping a 13-year-old girl make a suicide plan

    A couple in North Carolina, US, sued Google, accusing its flagship AI model Gemini of helping a 13-year-old girl make a suicide plan after she interacted with Gemini thousands of times on a Chromebook about suicide topics. Google says it is reviewing the accusations and claims to have introduced protective measures, but parents and schools criticize the promotion of AI features in its K-12 products.

  27. ITHome · AI40

    Honor 700 series phones go live on the official website: Pro / Standard / S models, all featuring back screen design

    Honor officially announces the launch of the Honor 700 series and opens pre-orders on the official website, with three models this time: Honor 700 Pro, Honor 700, Honor 700S, all featuring back screen design. The series phones use the only dual 8-core chip in the same tier, offering stronger performance, larger battery, periscope, flagship imaging, and selfie back screen, with a back screen size of 1.72 inches.

  28. ITHome · AI62

    Anthropic has not yet released an AI model to help solve major quantum computing challenges

    Anthropic's yet-to-be-released AI model helps the physical research institution PSI prove the Allen-Arnold conjecture, which has not been resolved since it was proposed in 2008 and involves when quantum computers can surpass the computational speed of traditional computers. The research system developed by PSI based on this model completed the proof in just three days, confirming that quantum advantage depends on special laws of the problem. Previously, OpenAI has used AI to solve the Navier-Stokes equation problem.

  30. ITHome · AI65

    Meta launches a dedicated version of existing models based on Muse Wonder training, with proactive memory and focus capabilities delegated

    Meta has launched a dedicated version of existing models trained on Muse Wonder, transferring proactive memory and proactive focus capabilities to the Muse and Meta AI dedicated model groups. Muse models have increased the proportion of correct use of key context in 90% of tasks to 76%, and the score of Meta AI models in Meta's 'Personal Super Intelligence Index' has increased from 38% to 47%. New models will be prioritized for subscription users, with subscription prices of $20 per month or $100 per month.

  32. ITHome · AI22

    Wccftech reports that the AMD Zen 6 sample scored, with a 10-core multi-core performance about 21.5% ahead of the previous 12-core generation.

    AMD Zen 6 10-core sample scored 3100 single-core and 18313 multi-core in Geekbench 7, 21.5% higher multi-core performance than the 12-core Ryzen AI 9 HX 370; the 8-core sample's multi-core score of 15075 matched the latter, while the 6-core sample's multi-core score of 10670 was 17% higher than the Ryzen AI 5 340.

  34. ITHome · AI65

    Apple's iOS 27.3 first developer preview adds a dedicated feature Flag, suggesting Siri AI localization for the Chinese market.

    Apple iOS 27.3's first developer preview version introduces a feature switch named LinwoodChinaMainland, associated with the IntelligenceFlow framework, which is the internal scheduling layer of Siri AI. This switch is currently disabled but is considered one of the necessary conditions for Chinese market devices to qualify for Siri AI usage.