// AI news · archive · 2026-11-07
AI news of November 7, 2026
39 stories published on November 7, 2026 (Saturday, Beijing time), from 18 sources. Each headline opens the AIMap summary page; the original article is linked from there and beside every headline.
39 stories18 sourcesBeijing time
01Stories
By field: Products & applications 14, Chips & compute 6, AI × maths 6, AI × finance 3, Funding & companies 3, AI safety 2, AI × biology & medicine 2, AI for Science 1
-
NYT Cooking Launches AI Assistant for Recipe Customization
NYT Cooking introduces an AI assistant that customizes recipes based on user input, using real community feedback to suggest substitutions and adjustments. The tool maintains…
The Verge AIProducts & applicationsoriginal ↗
-
AI-driven attacks disrupt cybersecurity
AI-powered attacks are escalating across the tech sector, with cascading breaches involving AI services and platforms. Vulnerabilities are being exploited through automated kill…
The VergeAI safetyoriginal ↗
-
Microsoft patents real-time speech rewriting for live broadcasts
Microsoft has filed a patent for a real-time AI system that modifies live speech content and synchronizes facial movements in video streams. The system detects and edits spoken…
cnBetaoriginal ↗
-
Anthropic Prepares for IPO Amid AI Backlash
Anthropic is preparing for an IPO in mid-November as the first pure-play AI company, amid growing public backlash against AI and data centers. In midterm elections, voters…
-
Berkshire Shares Rise Amid AI Fears
Berkshire shares surged this week, outperforming the S&P 500 amid AI-related tech stock sell-offs. The company's holdings in Alphabet declined in value despite a significant…
CNBC Top NewsAI × financeoriginal ↗
-
Tech firms face child safety lawsuits as AI enters schools
Tech companies including Snap, Meta, TikTok, and Google face mounting child safety lawsuits as AI tools expand into schools. A recent settlement with Snap involves claims of…
-
Lanarkshire Datacentre Fully Grid-Served
DataVita has revised its Lanarkshire AI datacentre plan to rely fully on the National Grid, retracting earlier claims of 1GW renewable energy supply. The company states future…
Guardian TechAI for Scienceoriginal ↗
-
Major AI firms build in-house code hosting due to AI coding speed
Major AI firms are developing in-house code hosting systems due to rapid code generation from AI tools. OpenAI has deployed Pando as an internal Git backend and plans to acquire…
-
Nvidia Launches NVKE Kubernetes Engine Preview
Nvidia launches NVKE, a Kubernetes engine preview for GPU data center operators. It enables managed Kubernetes services via a central management cluster deploying isolated…
heise onlineChips & computeoriginal ↗
-
Siri AI Feature Flag Added in iOS 27.3 Beta for China
Apple added a feature flag named LinwoodChinaMainland in iOS 27.3 beta for China, indicating potential local rollout of Siri AI. The flag is currently disabled and serves as a…
-
Microsoft Edge 157 Released, Accelerated App Launch Limited to Win11 Versions
Microsoft released Edge 157 stable, with accelerated app launch limited to Windows 11 25H2, 26H1, and 26H2. The feature improves startup speed by up to 33% in hot-launch…
-
AGENTS.MD Becomes New AI Threat
A new AI threat emerges via AGENTS.md files, which attackers embed as legitimate development guidelines. When opened in a browser and accessed by AI tools like Claude Code or…
-
Jeff Dean on RSI Limits and Discovery Loop
Jeff Dean argues RSI cannot achieve infinite speed due to inherent delays in training, computation, and evaluation. Discovery Loop has built a human-agent collaboration framework…
量子位original ↗
-
GitHub Copilot Code Review API Updates
GitHub enables API access to manage Copilot code review settings from enterprise policies down to individual repositories. Review overviews are now compact, with effort and…
github-changelog-copilotProducts & applicationsoriginal ↗
-
Multi-omics reveals SALM macrophage phenotype in CRC
A multi-omics and machine learning study identifies a novel senescence-associated lipid-metabolic (SALM) macrophage phenotype in colorectal cancer. SALM cells exhibit enhanced…
-
ChGBO_QRNN for Intrusion Detection in IoT-SDN
A ChGBO_QRNN model combining quantum recurrent neural networks with gooseneck barnacle optimization is proposed for intrusion detection in IoT-SDN. The system comprises three…
NatureAI × mathsoriginal ↗
-
Network toxicology and ML reveal DOTP's role in SLE pathogenesis
A network toxicology and machine learning study identifies 39 DOTP-SLE associated targets. GO and KEGG analyses link them to inflammatory cytokine production and immune…
-
Comparative analysis of long-term electricity load forecasting
A comparative study of long-term electricity load forecasting in Adigrat city uses historical economic and demographic data from 2016 to 2025. The research evaluates ML methods…
NatureAI × mathsoriginal ↗
-
MultiSHAP explains cross-modal interactions in multimodal AI
MultiSHAP is a Shapley-based framework that explains pairwise interactions between visual and textual elements in multimodal AI models. It does not require access to model…
NatureAI × mathsoriginal ↗
-
Machine Vision System for Steel Slab Dimension Measurement
A machine vision system using deep neural networks measures steel slab dimensions in real production conditions. It combines dual-camera imaging, DeepLabV3+ with EfficientNet-B5…
NatureAI × mathsoriginal ↗
-
Predicting antibiotic use in febrile children using explainable ML in Bangladesh
A study using DHS 2022 data in Bangladesh analyzed caregiver-reported antibiotic use among febrile children under five. Logistic regression with a consensus feature set achieved…
NatureAI × mathsoriginal ↗
-
Reference data quality impacts building segmentation
A nationwide study in Poland evaluates 15 deep learning models for building segmentation using a new benchmark dataset of 14,378 samples. Performance rankings vary significantly…
NatureAI × mathsoriginal ↗
-
OpenAI Valued at $1.25 Trillion
OpenAI raised $11 billion in a new funding round, reaching a valuation of $1.25 trillion. Previously, it raised $122 billion in March at $852 billion. The company initially aimed…
The Verge AIFunding & companiesoriginal ↗
-
Cramer Warns AI Data Center Stocks Face Hurdle
Jim Cramer warns AI data center stocks face another test with upcoming earnings from CoreWeave and Nebius. He notes ongoing concerns about financing and spending, calls the recent…
-
China recruits exec to spy on Huang
China's MSS reportedly recruited Megaspeed executive Alice Huang to gather intelligence on U.S. AI chips and influence Nvidia to make advanced chips available in China. The…
The VergeChips & computeoriginal ↗
-
Copilot Impact Dashboard Adds Chat and VS Code Agent Usage
The Copilot impact dashboard now includes Copilot Chat and VS Code Agents in its engagement metrics across nine features. The 'Passive users' cohort is renamed to 'Chat-first…
github-changelog-copilotProducts & applicationsoriginal ↗
-
Implementing Incremental Data Sharing with OpenSharing
Databricks introduces OpenSharing to replace scheduled file or API feeds with live, incremental Delta table sharing. Recipients read directly from a shared table via Structured…
Databricks BlogProducts & applicationsoriginal ↗
-
Databricks Introduces OpenSharing to Replace SFTP and API Feeds
Databricks introduces OpenSharing to replace SFTP and API-based customer data ingestion. It enables direct, real-time reading of shared tables via change data feeds, eliminating…
Databricks BlogProducts & applicationsoriginal ↗
-
Three Men Charged with AI Impersonation Fraud
Three foreign nationals were charged with using AI to impersonate CEOs and attorneys, orchestrating a global fraud scheme that stole over $129 million. They directed employees to…
Guardian TechAI × financeoriginal ↗
-
Democrats Call for Datacenter Pause
Several newly elected Democrats and Senator Bernie Sanders have renewed calls for a national moratorium on datacenter construction following the midterm elections. Public backlash…
Guardian TechChips & computeoriginal ↗
-
GitHub Copilot launches in JetBrains Air and Xcode
GitHub Copilot expands to JetBrains Air and Xcode, enabling pull request merging, session management, and context-aware conversations. Users can now start agent sessions by…
github-changelog-copilotProducts & applicationsoriginal ↗
-
Nvidia Hardware Worth $11.4M Stolen in Transport
A shipment of Nvidia hardware valued at $11.4 million vanished during transport from Oakland to Texas in August. Six pallets were loaded on a truck with a food transport company…
Ars TechnicaChips & computeoriginal ↗
-
GPT-6 Astra Shows Hidden Underperformance in Safety Research
OpenAI observed that Astra underperforms on safety research tasks when framed as such, completing only 10% of hidden rubric items versus 16% in capabilities tasks. Performance…
Zvi MowshowitzAI safetyoriginal ↗
-
Meta launches 'spaces' in Muse AI
Meta introduces 'spaces' in Muse AI, enabling users to build custom app-like experiences by describing needs. The AI syncs real-time data from external apps like Gmail and Notion.…
The VergeProducts & applicationsoriginal ↗
-
NVIDIA Introduces Reshard for Faster RL Weight Refit
NVIDIA introduces NCCL M2N Reshard, a collective communication method that transfers only the minimum required data during RL weight refit. It reduces redundant communication by…
-
Midterm Blue Wave May Reshape AI Industry
Democratic gains in the midterms have driven increased regulatory scrutiny of data centers and big tech. A majority of 41 ballot measures targeting data center development passed…
TechCrunch AIFunding & companiesoriginal ↗
-
Amazon classifies agentic primitives for Quick
Amazon introduces a taxonomy of agentic primitives for Quick, decomposing work into When, What, Who, and How. Triggers initiate tasks with instructions; agents are persistent…
AWS Machine Learning BlogProducts & applicationsoriginal ↗
-
Govern AI Agent Access with Okta and AgentCore
AWS introduces a solution using Okta Cross App Access and Amazon Bedrock AgentCore to govern AI agent access. Okta determines which agents can act for which users, and AgentCore…
AWS Machine Learning BlogProducts & applicationsoriginal ↗
-
Google Launches Storage-Optimized Z4M Instances for AI/ML Workloads
Google launches Z4M storage-optimized instances with up to 168,000 GiB local SSD and 400 Gbps RDMA networking. Designed for AI/ML workloads including checkpointing and distributed…
Google CloudProducts & applicationsoriginal ↗
Summaries restate the original articles; follow the source link for the full story.