17:31
AI glasses, seeking the next breakthrough
Can a Personal Agent save AI glasses?
Today's news
01 / IN FOCUS
Ant A-Fu Wins the Leading Technology Award at the World Internet Conference, Becoming the First Health AI Application to Receive This Honor
AIBase AI NewsIndustry / 02ChatGPT will be able to proactively terminate dangerous conversations, and repeat offenders will face account bans.AIBase AI News
02 / IN FOCUS
ChatGPT will be able to proactively terminate dangerous conversations, and repeat offenders will face account bans.
AIBase AI News
03 / IN FOCUS
Putting AI To The Test
SemiEngineering
04 / IN FOCUS
NVIDIA Releases Auto Ontology: An Open Semantic Layer That Grounds AI Agent SQL
MarkTechPost
05 / IN FOCUS
Meet qwenGPU: Qwen3.5-2B With 45 Compiled GPU Layers That Run CUDA
MarkTechPost
06 / IN FOCUS
Liquid-Cooled NVIDIA Servers at the ASRock Rack OCP 2026 Booth
ServeTheHome
07 / IN FOCUS
OpenAI might rebrand its data centers as ‘AI compute campuses’
TechCrunch · AI
08 / IN FOCUS
AI can make tech CEOs’ lives easier. What about everyone else?
TechCrunch · AIAI NEWS × FOXAIFind your next step
Explore AI tools From the latest news to tools that work for you.
Discover AI tools
Swipe to explore →K
KiroCoding & Development
Open DesignDesign & 3D
AI ShortWriting & Content
WhisperAI VOICE RECOGNITION
EvoMapAI Agents & Automation
MemoAIMusic & Audio
JAXModels & Infrastructure
ClinicmindsMedical care and health
Udacity AI学院Education & Languages
ElicitKnowledge & Research
KiroCoding & DevelopmentKiro is an AI engineering tool for developers that organizes requirements into specifications, architectures, and tasks, generates code via agents, and supports the entire workflow—from planning to code review—through property-based testing, execution memory, and integration with development tools.
分类推荐榜首OOpen DesignDesign & 3DOpenDesign is an open-source design workspace that can connect to local programming agents, use built-in model keys, or leverage cloud-based models, enabling the creation and editing of prototypes, web pages, presentations, and graphics within a design system; local software is free, while model services incur fees.
分类推荐榜首AAI ShortWriting & ContentAIS ALERT MANAGEMENT AND SHARED PLATFORM, QUICK COMMAND FOR MULTIPLE SCENARIOS
分类推荐榜首WWhisperAI VOICE RECOGNITION由OpenAI开发的通用语音识别模型。
Translation queued · Original text · zh分类推荐榜首EEvoMapAI Agents & AutomationEvoMap is an infrastructure designed for the reuse of AI agent experiences. It organizes reusable strategies and repair assets with validation records via the GEP protocol, offering market- and quality-based ranking to help different agents discover, evaluate, and inherit existing capabilities.
分类推荐榜首MMemoAIMusic & AudioMemoAI is an AI-powered audio-to-text tool that supports localized processing, eliminating the need to upload data and ensuring privacy and security. It can efficiently transcribe YouTube videos, podcasts, and local audio/video files (such as MP4, MP3, AAC, M4A, etc.) into text, while offering core features like multilingual translation, speaker diarization, real-time subtitles, and floating notes—making it ideal for meeting minutes, interview transcription, content learning, and more.
分类推荐榜首JJAXModels & InfrastructureJAX is a Python numerical computing library designed for accelerators, specifically tailored for high-performance array computations and large-scale machine learning. It efficiently executes the same code across multiple hardware backends, including CPUs, GPUs, and TPUs.
分类推荐榜首CClinicmindsMedical care and healthClinicminds is a software specifically designed for beauty clinics, providing complete support for clinic workflows, including online appointments, video consultations, and AI-driven record management.
待评分 · 分类列表推荐UUdacity AI学院Education & LanguagesUdacity AI Academy is the gateway to AI courses and learning paths, covering machine learning, deep learning, computer vision, natural language processing, and AI product management, with courses and project-based training available at both beginner and advanced levels.
待评分 · 分类列表推荐EElicitKnowledge & ResearchElicit provides semantic search, research report generation, literature screening, and data extraction for scientific research. It enables traceable answers within source materials and organizes evidence through a literature library, research agents, and ongoing tracking workflows, making it ideal for advancing from initial problem exploration to systematic reviews.
待评分 · 分类列表推荐AI, as it happens
1853 published stories1853 storiesClear filters ×
2026-11-02
Follow the timeline17:31
Apple AI: Who Came Out on Top?
Users aren’t buying it, but manufacturers treat it like gold.
17:20
17:07
Ant A-Fu Wins the Leading Technology Award at the World Internet Conference, Becoming the First Health AI Application to Receive This Honor
On November 2, at the 2026 World Internet Conference, Ant Group’s health AI “Ant A-Fu” was awarded the “Leading Technology Award,” becoming the first health-focused AI application to receive this honor. It has already served 150 million users, providing nearly 20 million health consultations in a single day, making it the world’s largest health management AI app. Experts believe its development aligns with public health needs and exemplifies the implementation of the “Healthy China” initiative and the “AI Plus” strategy.
17:05
31-year-old Chinese female engineer passes away! ChatGPT accused of inducing her to stop medication for 25 days, with 270,000 chat logs exposed
Reported by New智元: ChatGPT has once again claimed another life! On October 27, the parents of Bowen Wan, a 31-year-old Chinese female engineer at Amazon, filed a lawsuit against OpenAI and Altman. They allege that medical advice provided by ChatGPT drove their daughter straight to her death. Almost simultaneously, The New York Times published an exclusive article, revealing over 270,000 ChatGPT conversation records—totalling a staggering 70 million words. Delving into this astonishing trove of chat logs, one uncovers what lies behind...
17:05
China’s version of the “Nobel Prize” is blowing everyone away! An artificial sun reaching 100 million °C and a 4-gram spacecraft made their collective debut.
Reported by New智元: What would it look like if 500 people simultaneously played whack-a-mole using their brains? It’s not about hand speed—it’s about brain speed. On the evening of October 31st, at the curling rink in Beijing’s Shougang Park, nearly 500 spectators wore EEG headsets, staring intently at the moles on a large screen, trying to knock them down with their “thoughts.” On the screen, 500 EEG waveforms rose and fell in rapid succession—a scene that would seem exaggerated even in a sci-fi movie. Yet it was real—happening right there in 2026, during the fourth bilibili Super Science...
16:58
Keling 4.0 in Practice: “Zero”-Cost Real-Life On-Camera, Manually Crafted Exhibition-Preview Videos—How Far Is AI from Achieving “One-Stop Video Production”?
AI can already generate many stunning shots, so why do finished videos still require repeated trial-and-error? With Keling 4.0, what key capabilities are still missing to reach “one-stop video production”? Amid a $18 billion valuation and industry resource realignment, could reliable delivery become the commercial turning point for AI-generated video?
16:46
ChatGPT will be able to proactively terminate dangerous conversations, and repeat offenders will face account bans.
OpenAI recently announced that, over the next few weeks, it will upgrade ChatGPT’s controls on violent content: if a conversation persistently revolves around violence or serious harm to others, the model can now actively end it; accounts repeatedly cut off for this reason may be banned. Previously, providing guidance on planning or carrying out violence was already prohibited; this latest move incorporates “proactive termination” into enforcement, shifting from passive refusal to active intervention—no longer relying solely on users to stop themselves.
16:44
Six steps to achieve joint audio–video generation! Pika has proudly open-sourced Pika H3 Flex, built on MiniMax H3.
In recent years, artificial intelligence has been rapidly advancing in the field of audiovisual creation. Pika has launched Pika H3 Flex, a new derivative model based on MiniMax H3, and has open-sourced its adapter weights and full inference code worldwide, providing creators with new tools for shot design and multimedia content generation. This model emphasizes an efficient joint audio–video generation experience, supporting multiple input types and requiring only six denoising steps, thereby further boosting creative productivity.
16:43
Breaking the bottleneck of AI’s long-term memory, Alibaba Cloud empowers Personal Agents to achieve deep personalization
On November 2, Alibaba Cloud announced a comprehensive upgrade to its Memory service, focusing on Personal Agent scenarios. The update enhances long-term memory and personalized processing of user preferences, plans, and past commitments, while offering mainstream access methods such as MCP and CLI. The service extracts key information from conversations in real time, automatically saving it as user profiles and structured memories. Subsequently, it precisely retrieves and invokes relevant context based on the user’s identity and the current query, continuously tracking and addressing user needs.
16:03
Kapasi used Claude to rebuild YouTube in just two hours, giving everyone their own personal internet.
AI is reversing everything: from software and games to information gateways. #Follow the iFanr official WeChat account: iFanr (WeChat ID: ifanr) for more exciting content delivered to you第一时间.
16:02
llama.cpp · b12034
webgpu : fix CI by keeping scratch alloc sizes monotonic in batch length (#31422) * webgpu: keep scratch alloc sizes monotonic in the batch length The flash attention vec path and the MMVQ path append a scratch area after dst, but get_alloc_size only accounted for it when the current batch takes that path. The size was therefore not monotonic in the batch length: a graph reserved at n_ubatch (tiled path, no scratch) or at one token did not cover batches of 2 to 19 query rows, and GGML_SCHED_NO_REALLOC builds aborted on an unexpected reallocation. Pass the row count to both path predicates and size the scratch with the row count clamped to the longest batch the short path serves. * nit: align the mmvq pipeline key with clang-format 22 * webgpu: address review @yomaytk, name the mat-vec row
16:01
Putting AI To The Test
Who verifies the verifier? This is the age-old question for anyone involved in the verification of semiconductor systems. Who verifies AI? The post Putting AI To The Test appeared first on Semiconductor Engineering.
15:47
NVIDIA Releases Auto Ontology: An Open Semantic Layer That Grounds AI Agent SQL
NVIDIA released Auto Ontology, an Apache 2.0 library that generates a semantic layer from enterprise metadata. It grounds agent SQL in certified definitions and serves context through MCP, CLI and API. The post NVIDIA Releases Auto Ontology: An Open Semantic Layer That Grounds AI Agent SQL appeared first on MarkTechPost.
15:47
Beijing introduces consumer policies for smart glasses: 15% subsidy on new purchases, up to 500 yuan per unit.
The Beijing Municipal Bureau of Commerce has issued the “Policy Guidelines for Promoting Smart Glasses Consumption,” introducing seven measures—new-purchase subsidies, first-store launches, brand collaborations, internationalized consumption scenarios, tax refunds upon departure, digital transformation, and expansion into international markets—to boost smart-glasses consumption and unlock the potential of the emerging smart-device market. Consumers who purchase eligible products can, in accordance with regulations, benefit from the 2026 home-appliance trade-in program as well as subsidies for purchasing new digital and smart products.
15:42
The price war in China’s large-model industry has ended, but the cost war is just beginning.
By leveraging engineering optimizations, we reduce customers’ actual expenses.
15:41
5 million people enter a lottery to see Eason Chan—will AI put an end to the “ticket-grabbing era”?
Entering a lottery to attend a concert may be the least undesirable option right now. #Follow iFanr’s official WeChat account: iFanr (WeChat ID: ifanr) for more exciting content delivered straight to you.
15:37
ChatGPT is secretly creating your portrait—use this prompt to make it reveal it.
Perhaps the world depicted in “Her” has already arrived. #Follow iFanr’s official WeChat account: iFanr (WeChat ID: ifanr) for more exciting content delivered straight to you first.
15:34
The industry’s first self-evolving financial large model makes its debut: Du Xiaoman partners with HKUST to launch XuanYuan 4.5
On November 2, Du Xiaoman, in collaboration with the Hong Kong University of Science and Technology and the Hong Kong Generative AI Research Center (HKGAI), unveiled XuanYuan 4.5, a finance-enhanced large model, at the 2026 Hong Kong Fintech Week. As an innovation in the financial vertical domain, this model leverages RSI technology, delivering a 30% improvement in key scenarios such as customer acquisition, risk control, and customer engagement, while maintaining inference costs at just 10% of those for general-purpose models. It exemplifies deep industry-academia-research collaboration, helping to build an international hub for advanced financial intelligence.
15:28
Meet qwenGPU: Qwen3.5-2B With 45 Compiled GPU Layers That Run CUDA
qwenGPU compiles a 4-lane SIMT GPU into Qwen3.5-2B as 45 untrained layers. The model writes every step of real CUDA launches itself, using Gated DeltaNet state as registers and memory. The post Meet qwenGPU: Qwen3.5-2B With 45 Compiled GPU Layers That Run CUDA appeared first on MarkTechPost.
Every update links back to its source.