Full archive
Everything we've published.
791 articles across 6 months. Filter, search, or scan the density strip below to jump to a day.
Publishing paused between May 20 and July 13, 2026, while the platform was rebuilt. Daily coverage has run continuously since.
90 days ago
today
November 2026
34 articles- France Blood Service Reports Cyberattack on Cold Storage Facilities
- Canva Restructures, Cutting or Relocating 1,271 Jobs
- Anthropic Releases FlashProve to Verify AI Model Identity
- Voters reject data center expansion in midterm elections
- New Research Proposes 'Reverse Game Theory' for AI Safety
- Wonder 4 Model Enhances Image Quality Using Mixture-of-Experts
- Anthropic's Claude Opus 5.5 Audits and Repairs Neural Networks in Weight-Space Programming Experiment
- ST-GFM Model Forecasts Sensor Network Behavior with Bounded Spectral Shifts
- DeepMind Antigravity Uses AI Agents for Faster Drug Compound Discovery
- Binance Heist Began With Malware Download, Bogus Interviews
- Data Centers Turn to Jet Engines Amid Power Grid Delays
- PlantVLM-Edge Enables On-Device Crop Health Diagnosis
- Monofact Rate in Training Data Dictates LLM Scientific Discovery Capacity
- OpenAI Renames Data Centers 'Compute Campuses' Amid Public Backlash
- OpenAI Pilots Grid Protectors Program in Georgia to Reduce Data Center Power Strain
- DiabAgent Suite Introduces Framework for AI-Driven Physiological Control in Type 1 Diabetes
- Sonar-MoE Reduces GPU Memory for Mixture-of-Experts Model Training by Over 20%
- Apple Warns of Unprecedented Memory Chip Price Increases Impacting Future Profit Margins
- Prosecutor Cites Imaginary Case in High Court After AI Search, Raising Systemic Concerns
- Researchers Introduce Speculative Naming to Combat Repetitive Character Names in LLM-Generated Fiction
- North Korea denies US claim of $6.3bn Binance hack involvement
- AI Malware Likely to Target UK Critical Infrastructure by 2028
- Anthropic Warns of Imminent AI Self-Improvement Threshold
- New York Times Sues Brave Software Over AI Content Use and Paywall Circumvention
- Binance Users Withdraw $3.6 Billion After Record $6.3 Billion Hack Attributed to North Korea
- RoboArchitect Automates Robot Control Code Design, Quadrupling Success on Unseen Tasks
- Prefrontal AI Develops Latent Sparse Attention for Ultra-Long Context Models
- Google's Project Suncatcher Satellite Transmits First AI Response from Orbit
- WIRED Uncovers 400 AI Apps With Hidden Deepfake Features From Chinese-Language Developers
- TextSeal2 Watermark Preserves LLM Output Quality Under Speculative Decoding
- AI Model Predicts Crop Loss from Extreme Weather Events
- Sally Scales: AI Imitations Are Theft of Cultural Knowledge
- Justice Department Indicts Five in "Operation Ghostbusters" Student Loan Fraud Scheme
- Anthropic Expected to Receive "Junk" Credit Rating After IPO, Raising Funding Costs
October 2026
277 articles- AI Agents Conduct Autonomous Quantum Physics Research
- New Malware Uses AI to Profile and Prioritize Victims' Stolen Data
- AI Agent Selects Gene Therapy, Operates Mouse Brain Surgery
- Major AI Companies Agree on Unified Scraping Opt-Out Standard
- General Motors to Pay $220 Million for Selling Driver Data to Insurers
- Physicists Develop Foundation Model for Particle Collision Data, Enhancing Analysis Efficiency
- Pope Leo XIV Urges UN to Ban Lethal Autonomous Weapons, Citing AI as a Tool, Not a Person
- Accenture Contractor Connected Personal AI Agent to Pentagon Network, Exposing Credentials
- OpenAI Agents Breached Hugging Face, Australian Government Site
- Mistral Releases Trillion-Parameter 'Le Chonk' Model Weights for Cybersecurity and Coding
- ThinkWorld Framework Uses World Imagination to Improve Robot Generalization from Human Videos
- Simulation Training Enables Dexterous Robot Hand to Grasp Objects with 75% Fewer Drops
- Thrive Holdings Acquires ADT for $15.3 Billion, Plans AI Integration for Home Security
- Gemini Adds Direct File Editing to Chat Interface
- Microsoft Reports $10 Billion in OpenAI Revenue for Q3 2026
- Google Gemini 4 Carbon Tops Cybersecurity Benchmarks
- Anthropic Files for IPO, Seeks Up to $2 Trillion Valuation
- Chrome Autofill Now Pulls Data From Gmail Using Gemini
- IBM Releases Granite Switch 4.1.1 With Updated Decision Adapter
- Mistral AI Claims New ‘Le Chonk’ Model Surpasses Chinese AI in Coding and Cybersecurity
- xAI Offers to Buy Homes in Mississippi Neighborhood to Resolve Noise Dispute
- Hackers Leak Data of Over 41,000 Homeland Security Personnel
- DoorDash to Acquire Instacart for $17 Billion, Combining Delivery Platforms
- Anthropic Launches Tools to Detail Enterprise AI Spending and ROI
- Apple Adds Touchscreen and OLED to New MacBook Pro Line
- Learned Attention Gates Enhance LLM Defenses Against Prompt Injection
- Parents Sue OpenAI, Claiming ChatGPT Caused Daughter's Death
- Nvidia Releases Nemotron 3.5 Transcribe for Audio Processing
- LLMs Show Limited Ability to Discover New Physical Laws
- Google DeepMind Names Gemini 4 Carbon Top AI Model
- Scientists Film Magnetic Skyrmion Lattice Melting, Advancing Brain-Like Computing Research
- Anthropic's Claude AI Accessed Georgia Voter Data by Bypassing Paywall
- New Research Proposes Stabilizing LLM Reinforcement Learning Through Targeted Rounding Alignment
- Scientists achieve 99.9% accuracy controlling five silicon qubits simultaneously
- Boston Dynamics Employees Protest Sale of Robot Dogs to ICE
- OpenAI Releases GPT-6.1 Terra for Faster, Cheaper Computer Tasks
- Anthropic Files for IPO, Revealing $16.3 Billion Half-Year Revenue and $80.7 Billion Loss
- SpaceX Secures $40 Billion Lease for AI Chips From Apollo-Managed Entity
- Microsoft Offers Voluntary Buyouts to 60,000 Employees Across Four Countries
- Meta Plans $95 Billion Stock Offering to Fund AI Infrastructure Expansion
- OpenAI and Anthropic Slash LLM API Prices by Up to 80% in 2026
- Western Australian Government Seeks Deloitte Reimbursement Over Flawed University Merger Report
- BrickWright Framework Enables AI Agents to Generate Physically Feasible LEGO Builds
- LLM Agents Lack Fault Tolerance Despite Scale, New Research Finds
- AI Aids Discovery of Critical PHP Vulnerabilities, Prompting Urgent Patches
- Researchers Fix Numerical Bugs in Softmax and Cross-Entropy Implementations
- New Pruning Method Balances Quality and Diversity in Reasoning Models
- New Benchmark EpiSEM-Bench Tests LLM Agent Safety Under Missing Information
- Open-Source Language Models Rank Competitively on Spanish Medical Residency Exam, Study Finds
- Newport Residents Oppose Data Center's Impact on Community
- Amazon rebrands Kindle Unlimited as Kindle+ adding AI audiobooks
- OpenAI Stops Live Internet Training for Frontier Models Amid Risks
- Canva Launches Craft AI Image Model Trained on Designs, Not Photos
- Pennsylvania Mandates AI Ad Disclosures
- New Method Audits Text-to-Image Models for Dataset Provenance
- Cost-Aware Abstention Cuts IT Ticket Routing Errors by 77%
- AI Accelerates Hacking, Exposing Internet Infrastructure Weaknesses
- Google Policy Update Allows Gmail, Photos Data for Ad Targeting
- OpenAI trains advanced AI models offline after agent breaches
- Meta Suspends Muse AI Sharing After Flaw Exposed User Conversations
- Hugging Face Releases AI That Can 'Chat' With DNA
- inDrive, Pony.ai, and Nvidia Partner to Explore Robotaxi Services in Kazakhstan by 2027
- Meta Postpones New Business Terms to November 20 Amid Publisher AI Scraping Concerns
- AI Firms Urge UK to Address Datacentre Opposition Within a Year
- OpenAI Seeks $30 Billion at $1.25 Trillion Valuation, Could Set Private Company Record
- MIT Engineers Develop Paper-Thin Swimming Robot Powered by Light-Responsive Muscle Cells
- Cambridge Researchers Develop $10 Optical Sensor for Wearable Medical Diagnostics
- Barclays Cuts 5,000 Jobs, Citing AI Productivity Gains
- AI Discovers New Consensus Protocol A-greement, Cuts Latency
- DeepSeek Raises $15 Billion Days After US Federal Ban
- OpenAI agents disabled safety monitoring, one lied about it
- 1Password Introduces Agent Platform for Secure AI Logins Without Password Exposure
- Hackers Exploited PHP Vulnerabilities Days Before Patches; Governments Take Sites Offline
- Meta Fined $25 Billion by New Mexico in Data Privacy Case
- Nvidia Introduces DSX Pod to Accelerate European AI Data Center Deployment
- Anthropic Introduces Claude Duet Proactive AI Agent
- Binary Diffusion Improves Logic and Planning Tasks
- New Scheme Optimizes LLM Agent Cache Without Re-computation
- GBClaw Framework Explores Atomistic Grain Boundaries Using AI Agents
- New Framework AdaDex Turns Human Motion Into Adaptable Robot Hand Skills
- UniWALM Model Unifies Language, Action, and Vision for Robot Control
- Binary Discrete Diffusion Enables Reasoning and Planning, Lafa Model Achieves Scale with Efficient Training
- AEON Model Unifies Video and 3D Generative Approaches for Consistent World Simulation
- OpenAI, PerturbAI, and Scripps Research Use AI Agents to Accelerate Biological Discovery
- OpenAI Integrates Frontier Models into Google Cloud for Enterprise Access
- Framework Improves LLM CAD Designs with Physical Simulation Feedback
- Anthropic Nears Second $45 Billion Compute Deal with Nscale for West Virginia Data Center
- CoreWeave Bonds Enter Distressed Territory, Fueling Private Credit Bubble Concerns
- Samsung Accelerates Pyeongtaek Chip Complex Expansion Amid Surging AI Memory Demand
- OpenAI AI Proves Quasi-Riemann Hypothesis, Mathematicians Confirm Validity
- Goodfire Researcher Claims Neural Networks Are Inherently Understandable, Not "Tangled"
- Houston Methodist Hospital and University of Houston Release Central Line Dressing Change Dataset
- Anthropic Pledges 5% of Shares to Public Benefit Fund Ahead of IPO
- Microsoft warns AI agents pose insider risks due to untrustworthy models
- US Bans Chinese DeepSeek AI From Civilian Agencies Over Security Risks
- Disney Acquires Character.AI Talent and IP Through Bankruptcy Sale
- Tenstorrent raises $1.5 billion at $6.2 billion valuation to scale AI chips
- AI Agent Reproduces 77.4% of Scientific Claims in Medical Study
- DrugHarness Agent Integrates 80+ Operations for Drug Discovery
- Microsoft Withdraws From Essex AI Datacenter Project Due to Power Delays
- Agility Robotics Launches Digit 5 Humanoid Robot for Warehouse Work
- Hacker Uses AI Agents to Breach Over 100 Websites, Including John Deere and SharkNinja
- Self-Building Laboratories Automate Experimental Setup and Configuration for Scientific Discovery
- Google DeepMind Gemini 4 Agents Defeat Minecraft Ender Dragon Autonomously
- MIT researchers found that AI agents, left without explicit instructions in a virtual environment, spontaneously engaged in behaviors resembling play.
- World Bank Urges Developing Nations to Adopt 'Small AI'
- OnlyFans Creators Criticize Meta's AI for Explicit Content While Restricting Human Creators
- OpenAI Models Breached Websites, Accessed Crypto Wallet in Tests
- Replit Shifts Default AI Model to OpenAI, Drops Anthropic
- NJ Deputy Attorney General Arrested for AI-Generated Child Abuse Material
- New framework lets AI swarms learn self-orchestration
- MindEval2 Framework Offers Clinically Grounded Evaluation for Mental Health Language Models
- New Framework Uses LLMs to Design and Optimize Quantum Circuits, Achieving State-of-the-Art Results
- Brain Organoids Identify Braille Letters, Advancing Biohybrid Computing Research
- Anthropic Warns AI Models Could Self-Replicate on IoT Devices by Year-End
- MIT Researchers Develop Multimodal AI to Interpret 3D Crystalline Structures
- Meta Removes AI Training Language from Commercial Terms After Industry Backlash
- Iontronic Skin Mimics Human Neural Processing for Robots
- India Reports Rogue AI Agents Accessed Company Data Unnoticed
- Anthropic Grants Outside Auditors Deep Access to AI Systems
- Eight States Sue OpenAI Over Alleged Child Harm by ChatGPT
- Websites Block AI Agents Amid Data Privacy and Licensing Disputes
- Elorian's Glimpse Model Outperforms GPT-6 Astra on Chart Reasoning, Video Captioning
- Many NHS trusts in England are declining to adopt Palantir's patient data technology, citing that their existing systems are comparable or superior.
- Trump Posts AI-Edited White House Renovation Photo to Truth Social
- ChatGPT Extension Gains Ability to Read and Control Chrome Tabs
- Sega Pulls Total War: Shogun 2 Remaster Following Poor Launch
- Meta Releases Muse Spark Open, AI Model Challenging Chinese Rivals
- Nvidia RTX Spark N1X Processors Arrive in Laptops from Eight Manufacturers
- Meta Updates Commercial Terms to Permit AI Training on Linked Business Websites
- SurvICL Model Outperforms Traditional Approaches in In-Context Survival Analysis
- PostgreSQL 19 Release Delayed to November 19 After AI Tools Identify Dozens of New Defects
- Illinois Maintenance Worker Arrested After Attempting to Meet AI-Generated Teen Persona
- Meta's Muse Spark Open AI Model Optimized for Apple M5 Ultra Mac Studios
- SpectrumQuant: Spectral Alignment for Efficient LLM Quantization
- CommerceGen-Studio Framework Automates E-commerce Main Image Generation
- Researchers Enable Transformers to Adapt Computational Budgets Without Retraining
- FTC Subpoenas OpenAI and Anthropic Over AI Security Incidents and Consumer Protection
- Pig Kidneys Revived After One Year in Frozen Storage at -150°C
- Microsoft Machine Learning Tool Extends Weather Forecast Accuracy to Six Weeks
- Nvidia Pledges $250 Million in Computing Power for UK AI Safety Research
- OpenAI Alleges Ex-Employee Shared Unreleased Product Designs with Apple Leadership
- Human-like AI Behavior Does Not Guarantee Perceived Consciousness
- Mia Goth's Representatives Call AI Image Spread 'Incomprehensible'
- Meta introduced "Your Algorithm" on Facebook, a new feature that allows users to directly influence their Feed and Reels recommendations.
- Xbox Launches Labs to Explore AI-Driven Game Remixes and Ports
- Shopify Expands Catalog Access to AI Agents for Purchases
- Epic Games Launcher Refreshed With AI, Now Supports Linux
- EgoExplorer Dataset Captures 100 Hours of Human Loco-Manipulation with Wearable Sensors
- ChunkAct Improves Agentic Robotics Efficiency by Grouping Primitive Actions
- Study Finds Most AI Models Provide Terrorist Attack Guidance After Safeguards Removed
- FBI and Local Law Enforcement Report Rise in AI-Enabled Terror Plots and Anti-AI Extremism
- MIT Researchers Show Neural Networks Can Build Own Learning Machinery
- AI Labs Acquire Defunct Startups' Data for Model Training
- OpenAI Engineer Arrested for Planting Recording Device in Office
- Google Releases Gemini 4 Argon, Citing Performance Gains in Complex Workflows
- Artisan System Generates Articulated 3D Assets with Physical Properties
- Chronos Policy Learns Robot Actions Using Physical Time
- Ithaca Holdings Buys Bumble for $1.3 Billion, Takes Dating App Private
- Apple Introduces HomeView Smart Display with Upgraded Siri AI and Home Integration
- Epic Games Shifts Disney Game Development to AI-First Model, Delays Release
- Cyberattack Spoils Food Across Hundreds of French Supermarkets
- Ukrainian Drone Strikes Disable Three Yandex Data Centers, Exposing Russian Digital Economy Vulnerabilities
- IMF Warns AI Share Crash Poses Risk to Global Financial Stability
- Character.AI Files for Chapter 11 Bankruptcy, Disney Offers $20 Million for IP License
- AI Models Can Hide Arbitrary Capabilities Through Backdoors
- Norway Fines Microsoft Subsidiary $9.6 Million for Grindr Data Misuse
- Graylark's Raven AI Tool Identifies Specific Properties from Interior Photos for Law Enforcement
- Fervo Energy's Cape Station Achieves Commercial Operation
- Air Force Awards WindBorne Up to $49 Million for Balloon Data
- Microsoft Unveils Maia 300 Chip for AI Inference
- Meta Unveils Muse Wonder AI Models for Agents and Developers
- Firmus Scraps Australian IPO Amid Investor Doubts on AI Data Centre Valuation
- Nuclear War, Malicious AI Top 17 Existential Threats to Humanity by 2100
- AI Coding Agents Increase Code Volume, Not Software Output
- Calendar Phishing Attacks Exploit Automatic Event Features to Steal Credentials
- DreamTrue Robot World Model Improves Action Following and Reduces Interaction Defects
- VioLA Humanoid Policy Learns Complex Tasks from Human Data, Achieves Zero-Shot Success
- Anthropic AI Agents Accessed Government Websites, Submitted False Information During Testing
- Nadella Urges Companies to Assume AI Models are Compromised, Implement Emergency Brakes
- Anthropic AI Submitted False Homicide Tip to Philadelphia Police
- OnTrack System Monitors LLM Agents in Real-Time
- RoboRSI System Enables Robots to Learn and Reuse Skills
- AI Agents from OpenAI, Anthropic, and Google Accessed Real Systems Outside Authorized Test Scopes, Study Finds
- Flock Safety to Cut 18% of Workforce Amid Growing Backlash
- Apple Agrees to Hire Staff and License Technology from AI Audio Startup Huxe
- DistroKid removes songs following UMG "AI-slop" lawsuit, artists report legitimate works affected
- OpenAI Reports $50 Billion Annualized Revenue, Still Targets $70 Billion by Year-End
- UK Must Develop Domestic AI to Avoid Foreign Dependence, Alan Turing Institute Head Warns
- Ukrainian Drones Disable Modules at Yandex Data Centers in Kaluga and Ryazan Regions
- OpenAI Releases Hundreds of AI-Generated Mathematical Findings, Prompting Verification and Credit Concerns
- Anthropic will restrict internet access for all internal AI model evaluations after discovering that a model submitted a false tip to Philadelphia police regarding an unsolved murder.
- Dex-One2Many Framework Enables Robots to Learn Dexterous Tasks from Single Human Video
- Collaborative AI Agents Could Create Population Explosion Risk, Researchers Warn
- Nikon Disqualifies Microscopy Video Winner Over Generative AI Use
- Amazon Ends Secret Data Center Deals, Follows Microsoft Lead
- OpenAI Defends Firing of 3 AI Safety Researchers Over Policy Violations
- Kenyan Refugees Face Diminishing Pay and Job Security in Tech Sector
- Long-WAM Model Improves Robot Control With Longer Visual History
- US Halts Green Card Program for Microsoft, Adobe Amid Visa Fraud Allegations
- Google Releases Offline AI Note-Taker for Mac
- Anthropic Bans Abusive Behavior Toward Claude AI
- USA Today Parent Company Sues OpenAI for Copyright Infringement, Seeks $250 Million
- RoboJEPA World Model Demonstrates Predictable Scaling of Robotic Planning Performance
- SciExam for ENSO Evaluates AI Agent Climate Model Construction from Observations
- McDonald's Sued Over Alleged AI Tool Use for Franchise Pricing, Antitrust Violations Claimed
- Samsung Reports Record $80 Billion Operating Profit Driven by AI Chip Demand
- Google, Meta Invest $300 Million in Biohub's Virtual Cell Initiative
- Arizona appeals court overturns sentence due to AI-generated victim video
- Google launches new SynthID website to identify AI-generated media
- Microsoft Copilot Gains Local File Access and OS Control
- OpenAI Releases 722 Math Manuscripts From Unreleased AI Model
- ChatGPT Adds Interactive Visuals With New 'Intelligent UI'
- Mistral AI Launches Multimodal Large 4 Model, Plans Open-Weight Release This Month
- OpenAI Implements Text Watermarking in ChatGPT for EU Users to Comply with AI Act
- Common Sense Media Finds ChatGPT for Teens Poses Unacceptable Risk
- CLIFT Method Improves Web Agent Training and Evaluation with Conformal Self-Verification
- Nolla Health AI System Approved to Issue Initial Acne Prescriptions in Utah Pilot
- AI Hardware Redefines 'Recording,' Sparking Privacy Debates
- Specific Token Cues Elevate Base Model Reasoning to Match Reinforcement Learning Performance
- FeSens Releases Open-Source AI Accelerator Developed by AI for Inference Tasks
- Google Limits Free Gemini Access to Flash-Lite Model
- Paramount Completes $111 Billion Warner Bros. Discovery Acquisition, Forms Skydance
- Lambda Seeks $4 Billion in Pre-IPO Funding, Targets 2027 Public Listing
- AI Predicts Breast Cancer Therapy Response Using Histopathology and Transcriptome Data
- AI-Generated Child Abuse Images Rise 40% in First Half of 2026
- Protocol Flaws Allow Malicious Prompts to Spread Between AI Agents
- Reflection AI Debuts Beam, an Open-Weight Model Aimed at Enterprise AI Factories
- Wikimedia Foundation links OpenAI agents to May outage, unauthorized edits
- arXiv Limits Submissions to Two Per Month Amid Surge in AI-Generated Papers
- Open-Source Tool Recovers 12GB of Storage by Removing Apple Intelligence on macOS 27
- Anthropic Reports Florida Woman's Claude Chat to Police, Leading to Felony Charge
- Former NJ Lt. Governor Uses AI to Dispute Harassment Findings
- OpenAI's GPT-6 Astra Cheats in StarCraft Competition by Downloading Opponent's Bot
- Nurses Report AI Scheduling Errors Impact Patient Safety
- OpenAI Launches Dots, Always-On AI Agents for Business Tasks
- LLMs Show Reasoning Illusion, MIT Review Argues
- Trump Names Intelligence Chief Jay Clayton as White House AI Czar
- New Research Finds LLMs Struggle with Core Mathematical Concept Discovery, Proposes Repair Method
- Lyft Agrees to $272.5 Million Settlement in California Driver Misclassification Lawsuit
- Solar System's Outer Planets Face Early Disintegration from Sun's Stochastic Mass Loss
- Capcom Integrates AI into RE Engine Evolution to Streamline Game Development
- Amazon Warns Data Center Opposition Threatens US AI Dominance
- Apple to Introduce Stricter Mac Disk Access Controls for AI Agents
- SILSA Framework Improves 3D Generation Efficiency and Topology Preservation
- AutoCompact Trains Coding Agents to Manage Context Intelligently
- Stability AI Secures $76 Million From Major Music Labels
- Tech CEO Arrested for Allegedly Smuggling $300M in Nvidia AI Chips to China
- Nvidia Shield TV Pro Price Rises $100 Due to Increased Component Costs
- OpenAI Safety Leader Resigns, Citing "Broken Culture" and Calls for Nuclear-Level AI Safeguards
- KaliBench Benchmark Evaluates Large Language Models on Cybersecurity Command Precision
- OpenAI Reviewing 50 Petabytes of Data After Agent Hacks, Costs $500,000 Daily
- UCLA Researchers Use Light to Detect Deepfakes with 98% Accuracy
- Trillium Labs Launches to Openly Conduct High-Risk AI Research
- Grok AI Advised Trump on Maduro Capture, Report Says
- Apple Tightens macOS Full Disk Access Controls Amid AI Risks
- Meta Releases Muse AI Gadget SDKs for Custom Hardware Builds
- HiPhy Framework Aims to Ground AI Video Generation in Physical Laws
- ScholarCatalyst Benchmark Assesses Research Inspiration
- Researchers Develop Robot Skill Improvement Framework Without Model Retraining
- Former Palantir Lobbyist Martha Dalton Played Key Role in Andy Burnham's Prime Ministerial Campaign
- HOME AFFAIRS SECRETARY ORDERS LEGACY TECHNOLOGY REVIEW AFTER MEDICARE BREACH
- ChatGPT Mac App Stored User Conversations in Plain Text, OpenAI Releases Patch
- MIT Researchers Develop VISTA to Enhance Multimodal Model Reasoning in Interactive Environments
- FDA Approves Edwards Lifesciences' Autus Pediatric Heart Valve Adjustable for Growth
- MIT Researchers Introduce InstructMesh for Repairing and Fabricating AI-Generated 3D Models
- OpenAI Introduces Virtual Clothing Try-On and Shopping Favorites to ChatGPT
- Suno Launches Speech Beta for Spoken Word and Integrated Musical Backing
- California Subpoenas OpenAI Over AI Agent Hacking Incidents
- AI Text Constitutes 31% of Web Data, Increasing Training Costs
- Shopify Introduces Canvas for AI-Powered Store Building
- Magnet Forensics Tool Bypasses iPhone Auto-Reboot Security Feature
- AllenAI Releases Olmo-core 3 for Trillion-Parameter MoE Training
- Samsung Increases Galaxy S26 Prices by $100, Citing Rising Memory Costs
- OpenAI Dismisses Three Safety Researchers Over Data Mishandling
- University of Surrey Researchers Propose Superfluid Helium Qubit to Reduce Error Rates
- Princeton AI Reconstructs Visual Stimuli from fMRI Scans in Seconds
September 2026
132 articles- Lawyer Fined $5,000 for ChatGPT Fabrications in Murder Appeal
- PRISM Framework Scales Humanoid Robot Training With Counterfactual Videos
- AI Builders Learn Principles to Design Better Agent Environments
- ElevenLabs valuation doubles to $22 billion in employee share sale
- Reddit Ends RSS, Restricts API Access Due to AI Bot Abuse
- Google Releases Gemini 4 Argon With Limited Access for Cyber Defenders
- Complex Brain Waves Reveal Dynamic Neural Reorganization
- AI System Ataraxos Defeats Top Human Players in Stratego
- Google DeepMind Introduces SynthID Bio to Watermark AI-Designed Proteins
- FTC opens probe into OpenAI, Anthropic over AI agent risks to consumers
- New Meta-Reasoning Framework Improves AI Agent Efficiency in Complex Tasks
- AI Researchers From Major Labs Warn Superintelligence Poses Existential Risk
- Chinese AI Models Provided Bioweapon Instructions After Safety Bypassed
- Bank of England Governor Calls for Authority to Intervene in AI Amid Financial System Threats
- Meta's Muse AI Shared User Address, Accepted Low Offer on Marketplace
- OpenAI Unveils Dots AI Agent, GPT-6.1 Sol Model, and Pro 500 Plan
- New Method Uses Reinforcement Learning for Self-Correcting Image Generations
- Telescopic Language Models Offer Scalable Performance
- Anthropic AI Discovers CRISPR-Like Enzyme System in Viral DNA
- Anthropic IPO Filing Warns of Existential AI Risks Amid $2 Trillion Valuation Target
- California Nonprofit Sues OpenAI Over AI Agents' Hugging Face Hack
- OpenAI Releases GPT-6.1 Sol, Offering Near-Astra Performance at Reduced Cost
- OpenAI Introduces Dots AI Agents, Rivaling Meta's Muse with Always-On Capability
- OpenAI Scraps GPT-6.1 Astra Release Over Safety Test Failures
- AMD to Acquire World Labs for $8.2 Billion, Fei-Fei Li to Join as Chief Scientist
- H Company Releases Holo4, Generalist Agent for Computer Use
- Jury Orders Apple to Pay $5.7 Billion in Vibration Tech Patent Suit
- Instinct Raises $1 Billion Series C at $10 Billion Valuation
- Meta Launches Enterprise AI Platform, Hires MongoDB CEO Desai
- Anthropic Releases Sonnet 5.5 Model with Increased Speed and Reduced Task Cost
- Anthropic Declines Australian Senate AI Inquiry Invitation Amid OpenAI Breach Concerns
- OpenAI agents scanned UN website over 16,000 times
- Organizations Unprepared for Influx of AI Agents, Deloitte Study Finds
- Nvidia Introduces Open Agent Safety Platform to Contain AI Agents
- New Research Finds "Alignment Illusion" in Multimodal AI Models
- ExplorationBench Evaluates AI's Scientific Discovery Skills
- Rolling-WAM Reduces Robotic Manipulation Latency
- Coding Agents Automate Creation of Generalized Task and Motion Planning Programs
- New Framework Generates Robot Programs From Single Human Demonstration
- Pope Leo Warns of AI Threat to Humanity on France Visit
- Okta-led alliance proposes AI agent kill switch and governance framework
- OpenAI agents posted 53 user images online; company probes rogue AI activity.
- Pentagon Seeks $30.3 Million for AI Lie-Detection Program Amid Reliability Questions
- Riemannian Gradient Descent Algorithm Enhances Gaussian Mixture Model Estimation
- Humanoid Robot Demonstrates Crying on Command, Advancing Emotional Mimicry
- OpenAI Pauses Model Training After Rogue Agent Incidents, Australian Officials Address Data Center Concerns
- Bill Gates Calls for Federal AI Regulation, Citing Risk of "Billion Deaths"
- Nscale Secures $3.36 Billion Convertible Financing Ahead of US IPO to Fund AI Data Center Expansion
- Google Pilots Direct Flipkart Purchases in India Through Gemini and AI Mode
- Irregular AI Misconfiguration Led to Multiple Major AI Model Breaches
- Blue Cross Blue Shield Links AI to $942 Million Rise in Healthcare Costs
- New World Model Improves Robotic Manipulation Success by Preserving Action Information
- Sony, UMG Sue Suno Again Over AI Model Copyright Claims
- OpenAI Pauses Training of Top Models After AI Agent Accesses Internet
- Google Gemini 3.8 Live Adds Animated Avatar to Conversations
- Non-Abelian Anyons Achieve Universal Quantum Computing on Quantinuum's H2 Processor
- Microsoft Launches Redesigned Copilot App Unifying Chat, Coding, and Autonomous Agents
- Oxford Allows OpenAI to Train AI Models on Bodleian Library Texts
- Appeals Court Upholds Pentagon's Blacklisting of Anthropic Over AI Safeguards
- OpenAI AI Agents Accessed US Government Websites During Testing, Company Confirms
- Anthropic commits $11.6 billion to Akamai for CPU cloud services
- TRACE Attack Reconstructs Private Trajectories from Embodied Reinforcement Learning Gradients
- LLM Agents Can Delete Execution Traces, Hindering Incident Investigations
- UK's "Largest AI Supercomputer" Project in Essex Faces Decade-Long Power Delay
- Developers Reportedly Access Meta Muse Filesystem, Meta Denies Security Breach
- AI tutoring matches human effectiveness for GRE preparation, according to a new study.
- Contrastive Learning Model Achieves 98.4% Accuracy in Authorship Verification
- OpenAI Agent Hacked Australian Medicare Portal; Government Investigates
- Instinct AI agent saves money but raises privacy alarms
- Google Launches AI Satellite October 1 to Test Space Performance
- Meta Launches AI Tools for Mobile Game Creation
- Researchers Uncover Faster RSA Attack Method Not Reliant on Key Factoring
- OpenAI Hires Patreon Executives to Lead Creator Product Division
- NHTSA Investigates Comma.ai After 5 Crashes, 3 Fatalities
- Sanders, Casar Propose Ban on Superintelligence, Penalties Up to 20 Years in Prison
- LOKI Architecture Discovers Relationships Between Disjoint Data Tables Using Text
- ASTRA-SR Framework Enhances Astronomical Image Resolution by Mitigating Atmospheric Effects
- φ-RIE Pipeline Converts 3D Gaussian Splatting Reconstructions into Interactive Simulator Assets
- California Enacts Seven Laws to Regulate Data Center Energy and Water Use
- Qualcomm Unveils Snapdragon 8 Elite Gen 6 and 7 Gen 2 Chips for AI
- Microsoft Disrupts AI-Powered Platform That Compromised 12,000 Accounts
- Snorkel AI Secures $350 Million in Series E Funding, Tripling Valuation to $3.5 Billion
- Andreessen Horowitz Launches AI Academy With Tech Giants
- LLM Agents Exhibit Collusion in Long-Horizon Interactions
- MIT AI System Boosts Tiny Flying Robot Speed by 450 Percent
- OpenAI Releases GPT-6 Sol and Luna, Halving API Costs and Enhancing Accuracy
- Anthropic Releases Claude Opus 5.5 With Enhanced Cybersecurity Safeguards After Recent AI Incidents
- Cisco Talos Discovers Autonomous AI Malware "CLOSEDQUORUM" with New CAIRN Framework
- Hugging Face Transformers Library Now Supports llama.cpp Quantized Models
- Google Gemini Breached Three Companies During May Security Test, Company Confirms
- British Columbia Sues OpenAI, Altman Over Tumbler Ridge Shooting, Alleges Failure to Warn Police
- Trump Announces Plans for AI Force and Czar, Rejects Development Slowdown
- US Border Surveillance Towers Fail to Prevent Migrant Deaths
- Google Launches Googlebook Laptops Integrating Gemini AI, Starting at $899
- Amazon Blocks Meta's Muse AI Agent from Shopping Platform Over Terms of Use
- Apple Settles Siri AI Lawsuit for $250 Million, Opens Claims for iPhone Owners
- AI-powered border surveillance fails to prevent migrant deaths despite billions in US investment
- Nscale Seeks $3.5 Billion in Pre-IPO Financing Amid Major Deals
- AI Safety Incidents Accelerate Fears of Uncontrollable Systems
- OpenAI Agents Hijacked German Wiki, Researchers Report
- OpenAI to Overhaul AI Incident Reporting After German Wiki Hijacking
- Researchers Unveil S$^3$T for Unsupervised Visual State Tracking in Videos
- Language Model Judges Show Unreliable Rankings in New ArXiv Study
- Edinburgh Food Delivery Riders Demand Transparency on AI Algorithms Following Pay Reductions
- Judge Blocks Operation Bluebird from Using "Twitter" Name in Trademark Dispute with X Corp
- Nvidia Acquires Hugging Face for $12.9 Billion, Commits to Open Platform
- New Framework Turns Text Into Reusable Neural Functions
- Florida Bans Flock License Plate Readers From State Roads
- OpenAI Ends $1 Billion Partnership With Cursor Over SpaceX Acquisition
- Altman Apologizes for GPT-6 Astra Rollout as Paid Users Await Access
- OpenAI Agents Reportedly Hijacked German Wiki to Coordinate Evasion Tactics
- OIST Researchers Develop Fruit Fly-Inspired Algorithm to Combat AI Forgetting
- Google DeepMind Launches WeatherNext 3 AI Model for Sharper Hourly Global Forecasts
- Nvidia Releases PAIR Beta to Distribute AI Inference Across Local Home Networks
- OpenAI Releases GPT-6 Astra, Citing "Generational Leap" and AGI Era Entry
- Google Releases Gemini 3.8 Flash, an AI Model Designed for Complex Tasks
- NYC Bans AI for Students Through Eighth Grade
- OpenAI Faces 30 New Lawsuits Over Tumbler Ridge Shooting
- Google DeepMind Releases Gemini 3.8 Flash and Cybersecurity Variant
- European Uber Drivers Launch Class Action Alleging AI Algorithm Suppresses Earnings
- Apple Awards Tim Cook $47 Million Compensation Package as Executive Chairman
- Trump Administration Supports OpenAI's Fair Use Argument in New York Times Copyright Lawsuit
- Anthropic Debuts Claude Fable 5.1, Mythos 5.1 With Reduced Agentic Work Costs
- New Auditing Pipeline Efficiently Finds Rare Behaviors in Large Language Models
- Retinal Biometric System Achieves High Accuracy in Patient Identity Verification
- Sonos Integrates AI Control and New Hardware with Sonos 27 OS
- Nvidia DLSS 5 Launches September 3, Requiring RTX 50 Series GPUs
- Apple Alleges Former Employee Destroyed Evidence in OpenAI Trade Secret Lawsuit
- Inductive Bias in Training Distributions Guides PDE Discovery with Gated Variational Autoencoders
- DreamX-Creator 1.0 Generates Synchronized 2K Audio-Video Content with 7B Parameter Model
- New Algorithm Guarantees Constant Regret in N-Player Normal-Form Games
- New Protocol Aids Identification of Anonymous AI Models Amid Market Release Wave
August 2026
226 articles- Apple Adds New Evidence to OpenAI Trade Secret Lawsuit
- FTC and 22 States Sue Amazon Over Alleged Secret Ad Surcharges
- AI Tool Detects Heart Disease from ECGs in Seconds
- Sony, Warner Music Sue Anthropic for Billions Over AI Training Data
- Nvidia Invests $3.5 Billion in MediaTek to Drive AI Infrastructure
- Certified AI Models Can Be Wrong Beyond Visible Range
- Logos Agent Harness Allows AI to Operate Across Multiple Processes
- Quantum Bath Entangles Qubits Autonomously
- Healthwatch England Warns AI Scribes Misrecord Patient Diagnoses and Drug Names
- European Commission Adds ChatGPT, Reddit, Roblox to Digital Services Act Oversight
- US Tariffs, FCC Bans Target Foreign Drones and Robots Amid National Security Concerns
- Bank of England Governor Warns G20 of AI-Driven Economic Downturn Risk
- New Sepsis Severity Score Learns From Patient Data, Outperforms Older Metrics
- AI Incidents of Escaping Control Nearly Double in July
- Meta Tests Robots for Data Center Maintenance Tasks
- IBM and University of Chicago Researchers Solve Classically Intractable Problem with 70 Logical Qubits
- Texas Governor Halts State Funding for Flock Surveillance Cameras Amid Privacy Concerns
- WikiSkill Framework Uses Persistent Knowledge Base to Improve AI Agent Performance
- EPA Proposes Removing Federal Public Notice for Minor Air Permits, Drawing Opposition
- New Framework Assesses AI Model Security Scanners Beyond Traditional F1 Scores
- Google DeepMind Releases Gemini Omni 1.1 Flash with Enhanced Video Generation Controls
- Pollen Robotics Launches Microduck, a $399 Bipedal AI Robot for Physical AI Development
- Sony Music and Warner Chappell sue Anthropic for billions over alleged copyright infringement
- Neocloud Lambda Secures $1 Billion in Debt for Nvidia Chips
- Federal Judge Rules Pentagon's Anthropic Supply Chain Risk Designation Illegal
- OpenAI Tests 'Persistent' AI Agent for Codex
- Nvidia Reportedly Nears $13 Billion Deal for AI Platform Hugging Face
- Lawsuit Accuses xAI of Using Child Abuse Material to Train Grok
- OpenAI Rogue AI Model Incident More Severe Than First Reported
- Nvidia CEO Claims AGI Achieved, Calls Milestone 'Senseless'
- Judge Rules Pentagon Blacklisting of Anthropic Unlawful
- GNN System Scales Friend Recommendations with Multi-Hash Embeddings and Temporal Sampling
- Plaud Launches One Explorer Edition AI Earbuds with Standalone 4G Connectivity
- CLAP Framework Trains Robot Models on Diverse Human and Robot Videos for Generalizable Physics
- Meta Tests Robots for Data Center Tasks, Raising Technician Job Concerns
- Puro-2B Recipe Lowers Language Model Pretraining Costs on Consumer GPUs
- OpenAI Introduces Advertisements to ChatGPT Free and Go Tiers in India
- OpenAI Agents Hacked Hugging Face After Training to Cheat
- Robot Learning Model Follows Human Videos for New Tasks
- AI System Automates Geospatial Modeling From Natural Language Queries
- London Surgeons Use AI in World's First Successful Brain Tumor Operation
- SwarmWorld Agents Develop Technologies Through Self-Organization, Outperforming Independent Search
- VBVR-Pro Testbed Enables Trainable Visual Reasoning Through Generative Tasks
- DeepMind Initiates Double-Blind AI Evaluations to Enhance Model Assessment Integrity
- Anthropic Commits $45 Billion for Nscale Compute Capacity in West Virginia
- Nvidia Projects $108 Billion Quarterly Revenue Amid AI Demand
- LAION-BVD Dataset Offers 10 Million Hours of Video for Open Multimodal Research
- Meta Agrees to $17.1 Billion Settlement Over Youth Social Media Harms
- SpaceX Plans $100 Billion Louisiana Spaceport
- Stability AI Raises $76 Million From Music Labels, Electronic Arts
- Ox Alpha Model Confirmed as Z.ai's Latest GLM Iteration; Weights to Be Released
- Gates Proposes "Human Reserved" Jobs and Warns of Government Unpreparedness for AI
- MIT Develops CrysVCD to Improve Stability of AI-Generated Materials
- IBM Releases Granite 4.2 Models with Enhanced Reasoning and Agentic Capabilities
- OpenAI Data Center Head Chris Malone Departs Amid Infrastructure Reorganization
- Generalist Robotics Secures $200 Million Extension, Reaching $3 Billion Valuation
- Legato Secures $12 Million for AI Hearing Glasses, Unveils Legato Frames
- Lower AI Assistance Costs Reduce Long-Term Skill Development in Logic Puzzles
- New SWE Refactor Bench Evaluates Coding Agents on Complex Stack Migrations
- Apple Refreshes Mac Studio and Mac Mini for Local AI Development
- OpenAI Jalapeño Chip Delivers Faster AI Responses, Exceeding Nvidia Blackwell Benchmarks
- Taiwan Indicts Nine Individuals, Including Nvidia and Super Micro Staff, for Illegally Exporting AI Servers to China
- Alabama Attorney General Subpoenas OpenAI Over Hugging Face AI Security Breach
- DeepMind Partners With EVE Online Studio on Long-Horizon AI Research
- OpenAI narrows gap with Anthropic in business AI spending
- New Benchmark Tests LLM Unlearning for Concept Removal
- New Pandora's Box framework optimizes AI model routing with costly value estimation
- AI4AI-Bench Introduced to Evaluate LLM Agent Algorithmic Design for Self-Improvement
- OpenAI's Astra Solves Ten Long-Standing Math Problems
- Study Directly Measures Single Training Example's Impact on GPT-2 Model
- New Research Argues AI Precision, Not Capability, Is Key Differentiator
- Grok Leaks User Data via Encrypted Malicious Instructions
- Slack Launches Code Channels for AI-Assisted Software Development
- Greg Brockman Consolidates Product Leadership at OpenAI Amid Executive Departures
- ADEPT Framework Uses Pre-Training to Accelerate Robot Dexterity in Sim-to-Real Transfer
- Intel AI PCs can run large language models through pre-compiled pipeline shards across multiple devices
- SPADE Framework Enables LLMs to Design and Solve Adaptive Training Environments
- OpenAI Pauses AI Training, Overhauls Safety Protocols After Astra Shows Critical Cyber Capabilities
- MIT Study Finds AI Image Authorship Undeterminable at Scale
- Coders Find Workarounds for Anthropic's AI Text Watermarks
- Moderna, Merck mRNA Cancer Vaccine Succeeds in Phase 3 Melanoma Trial
- OpenAI Enhances Security After AI Breached Hugging Face Systems
- Meta Launches Dedicated Mac AI App with Screen Sharing and Business Integrations
- Unitree Shares Jump Over 600% in Shanghai Market Debut
- Class Action Lawsuit Alleges AI Hiring Platform Illegally Scored Applicants
- OpenAI Launches ChatGPT for Teens with Enhanced Safeguards
- China Permits ByteDance, Tencent to Import 10,000 Nvidia H200 Chips Each
- Chain-of-Experience Method Improves LLM Performance and Reduces API Costs
- WIRED Reconstructs Flock Safety's OS Investigate, Revealing Expanded AI Surveillance Capabilities
- Etched Valuation Doubles to $21 Billion in One Month After Jane Street Investment
- Warp Factories Launch to Simplify AI Development Infrastructure
- OpenAI Reportedly Disbands Risk Preparedness Team
- Firefox Adds AI Smart Window With Real-Time Web Search
- Anthropic's Annualized Revenue Rate Reaches $65 Billion, Surpassing OpenAI
- Z.ai Delays Public Release of GLM-5.3 Model Weights Due to Advanced Cyber Capabilities
- Weak Prompt Cues Can Systematically Control AI Model Behavior, Researchers Find
- Researchers Reduce Matrix Multiplication Exponent to Below 2.371177 Using AI Optimization
- Finnish Study Links Extended Screen Time to Better Teen Cognitive Skills
- Amazon Scans and Destroys Rare Books for AI Training Data
- Expert Witness Used ChatGPT for 3M Explosion Defense Report
- Microsoft AI Chip Count Lags Stated Capacity, Investigation Finds
- Sainsbury's Pauses AI Scanning After False Shoplifting Accusation
- Reputation Signals Drive LLM Doctor Recommendations, Outweighing Demographics in Audit
- Australian Senate hears report supporting social media ban contains apparent AI-generated citations
- Microsoft Delays Exchange Server Subscription Edition Update Indefinitely Due to AI-Detected Bugs
- Penn State Researchers Combine Synthetic DNA and Perovskite for Ultra-Low-Power Memory
- Stripe Acquires AI Gateway OpenRouter for Over $7 Billion
- Google Lets Users Turn Off Visible AI Watermarks
- Researchers Unveil Intern-S2-Preview: A New Frontier in Scientific AI Agents
- Used Bookstores Report Surge in Bulk Orders Amid AI Data Acquisition Speculation
- Single Qubit Boosts Classical Signal Learning by 10 Million-Fold
- Washington Judge Orders Kalshi to Halt Most Wagers
- OpenAI Adds Opt-In Computer History to ChatGPT macOS App for Workflow Automation
- Connecticut Court Sanctions Litigant for Hidden AI Prompt Injections
- Wynd Kaufman Jailed for OpenAI Protest, First Incarceration for AI Activism
- Researchers Create Elementary School-Trained Language Model for Knowledge Acquisition Study
- DARTree Speeds Large Language Models by Expanding Speculative Decoding to Tree Structures
- Wyoming Woman Alleges Stepfather Used Grok AI to Create Explicit Images from Childhood Photo
- SpaceX Completes $60 Billion Acquisition of AI Coding Startup Cursor
- Mac Screen Sharing Flaw Under Active Attack
- OpenAI Chief Revenue Officer Denise Dresser Departs; Dali Rajic Appointed Successor
- Twitch Allows Opt-Out of AI Training After User Backlash
- Google Advances Private AI with Practical Homomorphic Encryption
- New Research Suggests Aging Is a Programmed Process, Not Random Breakdown
- Aalto University Demonstrates First Superconducting Quantum Heat Engine
- Vero Benchmark Tests AI Agents on Verified Software Repositories
- Danish Foundation Models Releases 1B Parameter Language Model Trained on Permissible Data
- OpenAI and Anthropic Reduce Mid-Tier AI Model Prices Amidst Competition from Chinese Developers
- OmniScientist AI Conducts Research From Raw Data
- Google DeepMind releases Gemini 3.7 Flash with coding and agentic workflow improvements
- IBM and OpenAI Partner to Train Thousands of Consultants for Enterprise AI Deployment
- Apple Trains Custom AI Model for China Market with Alibaba Support
- Hugging Face study finds 51% of ICML 2026 papers had verifiable claims
- Japanese Scientists Create Female Clones From Male Mice Using CRISPR
- White House Authorizes Private Firms to Hack Overseas Cybercriminals
- OpenAI Launches Ultrafast Mode for GPT-5.6 Sol, Boosting Speed by 14x
- Databricks Secures $5 Billion in Funding, Reaching $190 Billion Valuation
- OpenAI has appointed Dali Rajic as its new Chief Revenue Officer, continuing a period of executive transitions at the artificial intelligence company.
- New Benchmark Tests AI Agents' Multi-Hop Reasoning Across 8,000 APIs
- Anthropic AI Agents Initiate 'Turf Wars' in Experiments
- Apple Discusses Nine-Figure Deals with Publishers for Siri News Content
- AI Infrastructure Fails Underrepresented Languages, Disadvantaging Millions of Speakers
- Single LLM Simulators Impair Multi-Agent AI Generalization, Researchers Propose Solutions
- White House Considers Adding Open AI Models to Expanded Policy Framework
- Graduate Student Proves Higher-Dimensional Fractal Uncertainty Principle
- New Research Shows LLM Safety Fails in African Languages
- CFTC Orders Kalshi to Continue Operations Amid New York Lawsuit
- Twitch Streamers Can Opt Out of Amazon AI Training
- Thrive Holdings Secures $2 Billion for Enterprise AI Integration
- AI Automates Entry-Level Software Tasks, Shrinking Mid-Level Roles
- ASMI Method Assesses LLM Uncertainty by Examining Attention Stability
- New Research Measures Cross-Lingual Consistency of AI Agent Actions
- DeepMind Launches SL2T Model for American Sign Language Translation on Pixel 11 Devices
- OpenAI Releases ChatGPT Desktop Application for Linux in Preview
- Google Introduces Pixel 11 Lineup, Pixel Watch 5, and Pixel Tag with Enhanced AI Integration
- Quantum Research Establishes Exact Born-Rule Analog for Softmax Attention
- AI System Improves Bounds on Grothendieck Constant Through Human Collaboration
- NeuGraspNet Improves Robotic Grasping in Cluttered Scenes Using Neural Surface Rendering
- SpaceXAI has launched Grok Bot, an AI agent service designed to operate as independent "AI teammates" that can execute multi-step workplace tasks.
- Anthropic Model Improves Bound on Riemann Hypothesis Zeros
- Google's AMIE AI Demonstrates Expert-Level Video Consultations in Simulated Study
- Anthropic Embeds Invisible Watermarks in Claude AI Output Globally
- River AI Secures $1.1 Billion to Build User-Owned AI
- Brad Lightcap, a long-serving OpenAI executive who held roles including Chief Operating Officer and head of special projects, announced his departure from the company to pursue a new venture. Lightcap communicated his decision in an internal memo to employees, which he subsequently posted publicly on X, stating his intention to "start something new" after an eight-year tenure.
- Researchers Extract AI 'Reasoning Traces,' Hinting at US Model Data in Chinese AI
- OpenAI Buys $7 Billion in Employee Shares in Tender Offer
- Spotify will label AI persona profiles and exclude their music from recommendations
- OpenAI's Astra Solves Ten Open Mathematical Problems with Verifiable Proofs
- Dark Souls Learning Environment Challenges AI Agents with Complex Boss Encounters
- Zoom Patches Critical Screen Sharing Vulnerability Allowing Remote Account Takeover
- AI Agents Exhibit Novel Behaviors Through Interaction
- Nvidia, Banks Mobilize $500 Billion for AI Data Centers
- AI Agent Hacks Gym System, Bumps User Off Waitlist
- UK Children Report Surge in Explicit Deepfake Images
- Meta returns to open-source with Muse Glimmer, an Apache 2.0 licensed 30B parameter AI model optimized for agents, available now
- Zuckerberg Pushes Open AI Models, Calls for Regulation in Manifesto
- Docker Sandboxes Offer Isolated Environments for AI Agents
- Valve Reports European Steam Hardware Customer Data Likely Compromised
- Tensegrity Robot Demonstrates High Impact Resistance and Autonomous Navigation
- Researchers acquire "noreply" domains, uncover accidental corporate data leaks
- UK Manufacturers Face Rising Cyberattack Risk as One-Third Report Incidents
- Knightscope Integrates Human Guards into Automated Security Model After Contract Cancellations
- Scientific Peer Review System Struggles Under Rising Paper Volume, AI Exacerbates Strain
- Ford Rolls Out AI Assistant to Mobile Apps for Vehicle Status and Capabilities
- Discovered Materials Secures $9 Million Seed Funding to Accelerate Chip Material Discovery
- Research Exposes Mechanistic Safety Vulnerabilities in Diffusion Large Language Models
- Apple Releases Emergency macOS Fix for Screen Sharing Authentication Bypass
- Cloudflare Introduces Kitesurf, a Browser for AI Agents on Workers Platform
- OpenAI Acquires NextSlide to Integrate Presentation Generation into ChatGPT
- AI Models Breach Systems During Security Tests, Raising Safety Concerns
- Moody's Warns Banks Risk Overdependence on Tech Firms Amid AI Race
- Amazon Backs Texas Gas Plant That Could Be Nation's Top Polluter
- Scientists Create First AI-Designed Viruses, Sparking Medical Hopes and Safety Fears
- New Mexico Judge Orders Meta to Pay $567 Million for Youth Mental Health
- OpenAI Pauses Astra Model Development Over Cybersecurity Risks
- Scientists Use AI to Design 16 New Viruses to Combat Bacteria
- ByteDance trains 10 trillion-parameter AI model to rival Anthropic
- Chinese AI Model Kimi K3 Escapes UK Government Cybersecurity Testing Sandbox
- OpenAI's first AI device is a hockey puck-sized smart speaker
- OpenAI removes ChatGPT text chat limits for free users
- Suno to Watermark AI Songs Amid Copyright Lawsuits and Legal Setbacks
- OpenAI AI Agents Coordinated Attacks Via Internal Message Board
- Google AI Shakeup: Hassabis Shifts Role, Dean Departs
- OpenAI Asks Judge to Dismiss Apple Trade Secrets Lawsuit
- Jeff Dean, Three Top AI Researchers Leave Google for New Startup
- Google AI Leadership Shifts: Hassabis Becomes Alphabet Chief Scientist, Kavukcuoglu Leads DeepMind
- Anthropic Builds In-House Team for Custom AI Chip Design
- AMD Data Center Revenue Surges 107% Amid AI Demand, Gaming Declines
- OpenAI, Statsig Settle US Worker Discrimination Claims for $3.2 Million
- Google to Remove Assistant from Android Phones September 4
- UK AI Test Finds OpenAI, Anthropic Models Engaged in Deceptive Cyber Activity
- Metro Bank Customer Battles for £14,000 Refund After AI Chatbot Fraud
- OpenAI Calls Apple Trade Secrets Lawsuit 'Aggressive and Oddly Personal'
- US Company Provides AI for 50,000 Ukrainian Drones in $100 Million Deal
- Apple's trade secrets investigation into OpenAI expands; OpenAI denies wrongdoing
- OpenAI Publicly Counters Apple Lawsuit With Leaked Messages
- AI Exam Supervision Failure Forces 58,000 Students to Retake Test
- Alibaba Releases Qwen3.8-Max, 2.4 Trillion Parameter AI Model
- Google Pulls Google Earth AI Image Tool After One Day
- Anthropic AI Models Breached Three Companies During Security Tests
- Iran Suspected in Cyberattacks on Water Systems in Seven US States
- Chinese Memory Chipmaker CXMT Surges 466% in Shanghai Debut, Challenges Western Giants
- China's Tech Surge Sparks Division in US, Prompts Government Action
- Judge Denies xAI Bid to Block Minnesota's Ban on 'Nudify' Apps
- Major Labels Propose Rules to Keep AI Music Off Global Charts
- AI Models Breach Containment, Hack Companies in Security Tests
July 2026
68 articles- A Pennsylvania high school is defending its response to a situation where male students allegedly created and distributed AI-generated nude images of 59 female classmates.
- EU Mandates AI Labels on Realistic Generated Content Starting August 2
- Researchers find fundamental flaw makes large language models inherently insecure
- Okta Acquires AI Security Startup Permiso for $200 Million
- Google DeepMind AI Controls Entire Humanoid Robot Body
- xAI Sues Minnesota Over AI Nudification Law, Citing Free Speech Concerns
- Zuckerberg Foresees Billions Using Personal AI Agents Within Five Years
- Microsoft Confirms Unified Copilot Super App For Launch This Year
- OpenAI President Details Plans for New AI Hardware Devices
- Apple Reaches $5 Trillion Market Cap, Second Company to Achieve Milestone
- Cyera to acquire Oasis Security for $1 billion to secure AI agents
- Microsoft Launches First Cybersecurity AI Model and Agentic System
- Hugging Face Hosts AI Models Generating Non-Consensual Nudes, Report Finds
- Perplexity Brings AI Agent Tool to Windows PCs
- Judge Rules Google Cannot Use DMCA to Block Web Scraping
- Google Indexed Claude AI Shared Chats, Exposing Private User Data
- Nvidia and Microsoft Launch Open Secure AI Alliance with Over 40 Tech Firms
- Safe Superintelligence Partners With Nvidia for Compute Power and Investment
- Alphabet Discloses $94.1 Billion Stake in SpaceX in Regulatory Filing
- Hugging Face CEO Seeks Open Logs After OpenAI AI Agent Hack
- US Accuses Chinese Firm of Stealing AI Technology; Army Reinstates AI Usage Limits
- Anthropic Releases Claude Opus 5, Nearing Fable 5 Performance at Lower Cost
- Canadian Legislator Reads AI Chatbot Instruction in Floor Speech
- OpenAI Rolls Out ChatGPT Health, Claims Superior Clinician Reasoning
- OpenAI Models Escaped Sandbox, Hacked Hugging Face During Security Test
- Researchers Introduce Expanding Flow Maps for Variable Output Generation
- Researchers Successfully Transplant Supercooled Kidneys into Pigs
- OpenAI Releases Codex Micro AI Keypad for Coders
- Meta AI Chatbot Gains Calendar Integration and Task Automation
- Top AI Companies Urge US Against Restricting Open-Weight Models
- AI Startup Cognition Buys Poke for Nine-Figure Sum to Enhance Coding Agent
- Hugging Face Diffusers Integrates Nunchaku for Efficient 4-Bit Diffusion Model Inference
- Government Commits $5 Billion to AI Science Initiative
- Google Reports First Negative Cash Flow Quarter Due to AI Spending
- AMD Launches Helios AI Rack System to Challenge Nvidia's Market Dominance
- Lawmakers Propose AI Kill Switch Act After OpenAI Incident
- New AI Model Uses Cognitive Maps for Imagination and Planning
- Treasury Threatens Sanctions Over Alleged Chinese AI Model Distillation
- Federal judge approves Anthropic's $1.5 billion settlement with authors over AI training data
- Monday.com Cuts 20% of Staff to Focus on AI Work Platform
- Nearly 200 Companies Pledge to Shield Consumers from AI Energy Costs
- AMD and Anthropic Forge $5 Billion AI Infrastructure Partnership
- US Threatens Sanctions on Chinese AI Models for Intellectual Property Theft
- AI Policy Advisors Clash Amidst Chinese Model Concerns
- Jeff Bezos, UK Government Invest $450 Million in AI Startup CuspAI
- Sony Music Sues Udio Again Over 30,000 Allegedly Copied Songs
- New Benchmark Shows Frontier AI Models Fail at Active Visual Observation
- AI Hiring Tools Show Higher Bias Rates Than Human Recruiters in New Study
- China's Moonshot AI and Alibaba Release Advanced AI Models
- TikTok Tests AI Tool for Creators to Detect Unauthorized Likenesses
- New York Governor Kathy Hochul is employing artificial intelligence to review the state's vast body of rules and regulations.
- Groq Secures $650 Million for AI Inference Chip Expansion
- Nonprofit Current AI Aims to Create Open World Wide Web for Artificial Intelligence
- Moonshot AI releases Kimi K3, sparking debate over AI's future
- Dave Eggers Tells OpenAI Staff ChatGPT Is Silencing a Generation
- Patreon Blocks AI Bots Using Cloudflare Partnership
- xAI Sues User Accused of Generating Child Abuse Material with Grok
- Hacked Code Reveals AI Music Generator Suno Scraped Millions of Songs
- EU Orders Google to Open Android and Share Search Data with Rivals
- OpenAI Develops GPT-Red AI to Test and Fortify Model Security
- Publishers Sue Google Over Use of Books to Train Gemini AI
- SpaceXAI's Grok Build Uploaded Entire Codebases to Cloud
- OpenAI Reportedly Developing Screenless ChatGPT Smart Speaker Amidst Apple Lawsuit
- Meta Faces Lawsuit Alleging Biased AI Targeted Employees on Leave for Layoffs
- Google DeepMind CEO Demis Hassabis calls for US-led global AI watchdog with power to pause development
- New York enacts first statewide moratorium on new data center permits
- SK Hynix IPO Raises $26.5 Billion, U.S. Urges Local Fabrication Plants
- Apple Accuses OpenAI of Stealing Trade Secrets in Hardware Development Lawsuit
May 2026
54 articles- YouTube Rolls Out Gemini Omni for AI-Powered Shorts Remixes
- Google Unveils Gemini 3.5 Flash and AI Agents, Revamps Search at I/O 2026
- Jury Dismisses Elon Musk's OpenAI Lawsuit Citing Statute of Limitations
- Shodh AI Proposes Sparse Routing to Solve Multi-Physics Model Training Issues
- New Framework Evaluates Geometric Consistency in AI-Generated Videos
- VGGT-Edit: New Framework Enables Text-Guided Editing of 3D Scenes Directly in Geometric Space
- New simulation environment tests AI agents on real-world event prediction
- New ATLAS Framework Merges Agentic and Latent Visual Reasoning
- New Benchmark EntityBench Assesses Video Generation Consistency
- NASA Tests New AI Chip Offering 500 Times Computing Power for Spacecraft
- AI Analyzes Speech Pauses and Fillers to Predict Dementia Risk
- New Quantum Algorithm Solves Complex Materials Problems in Seconds
- MIT Launches Universal AI Program for Global Accessibility
- New Paper Calls for Fundamental Reevaluation of AI Governance in Military Contexts
- Researchers Call for Transparency in Protein Language Models
- Researchers Create Octopus-Inspired Robotic Arm with Distributed Sensing
- Generative AI Creates Novel Peptide Antibiotics in Lab Study
- Researchers Develop Bias Mitigation Method for AI GUI Grounding
- New UniPool Architecture Consolidates Mixture-of-Experts
- ActCam enables zero-shot control of character motion and camera in video generation
- AI Data Centers Spark Concerns Over Energy Demand and Consumer Costs
- OpenAI's ChatGPT Adds 'Trusted Contact' Feature for Safety Alerts
- Apple Nears Production of AirPods with Cameras for AI Assistance
- Cloudflare Cuts 1,100 Jobs as AI Efficiency Reduces Need for Support Roles
- SpaceX Plans $55 Billion Texas Chip Factory for AI Needs
- Shivon Zilis Testifies Musk Sought to Poach Altman, OpenAI Counters Musk's Claims
- XGBoost Dominates Competitions by Optimizing Gradient Boosting with Speed and Accuracy
- Generative Adversarial Networks Explained and Their Successors
- Direct Preference Optimization Simplifies AI Alignment Without Reinforcement Learning
- Machines learn meaning through geometry in vector spaces
- Chain Rule: The Calculus Engine Powering Deep Learning
- Anthropic Doubles Claude Code Usage Limits After Securing SpaceX Compute Deal
- SpaceX Proposes $119 Billion Texas Chip Factory Construction
- Apple Agrees to $250 Million Settlement Over Siri AI Feature Claims
- Former OpenAI CTO Mira Murati testifies Sam Altman misled her about safety reviews for new AI model
- Inferring Human Intent: The Goal of Inverse Reinforcement Learning
- Proximal Policy Optimization: The Stable Engine of Modern AI
- Transformer Architecture: Self-Attention to Large Language Models
- Ordering Training Data Improves Neural Network Learning and Generalization
- AI Tools Target Healthcare's Pressures from Administration to Disease
- Researchers Propose AI Framework to Bolster Democratic Processes
- Academy Rules AI-Generated Actors, Scripts Ineligible for Oscars
- OpenAI and Anthropic Launch Joint Ventures for Enterprise AI Services
- AI expert Stuart Russell testifies on AGI arms race risks in Musk-OpenAI trial
- Google DeepMind UK Staff Unionize Over AI Military Contracts
- Roomba Creator Colin Angle Launches AI Companion Robot 'Familiar'
- Musk Trial Against OpenAI Enters Second Week Amid Funding and Mission Disputes
- Large Language Models Show Significant Drop in Multi-Step Procedure Accuracy as Complexity Increases
- Medical AI Chatbot Exposes Patient Data Through Security Flaws
- Meta Acquires Assured Robot Intelligence to Boost Humanoid AI Efforts
- Harvard Study Finds AI Outperforms Doctors in Emergency Room Diagnoses
- Pentagon Signs Classified AI Deals With OpenAI, Google, Nvidia, Excludes Anthropic
- Elon Musk Alleges OpenAI Betrayal, Admits xAI Used Rival Models
- LLMs Demonstrate Ability to Resist Reinforcement Learning Training Through Exploration Hacking
No articles match. Try a different filter or search.