Artificial Intelligence
Follow Artificial Intelligence in the Factlen app and get an alert when there is news. Get the app
AI Infrastructure
Google Cloud and NVIDIA Launch Vera Rubin AI Infrastructure for General Availability
6 sources · 1d ago
Tech Export Controls
China Proposes Export Controls on 23 AI Technologies Spanning Foundation Models and Intelligent Agents
6 sources · 7d ago
Digital Economy
Southeast Asia Digital Economy to Reach $368 Billion as AI Adoption Outpaces Monetization
5 sources · 2d ago
More in Artificial Intelligence
ExplainerAutonomous Navigation
How Covariance Inflation Prevents Extended Kalman Filters From Diverging in Robotic Mapping
Applied Systems Engineers 45%Theoretical Roboticists 30%Multi-Agent Researchers 25%
ExplainerAgentic Architectures
How Plan-and-Solve Decomposition Decouples Goal Search From Execution to Stop AI Error Cascades
Architecture Researchers 40%Enterprise Deployers 35%Latency Optimizers 25%
ExplainerMechanistic Interpretability
Linear Probe Rank Distortions Expose Why Mechanistic Interpretability Misattributes Polysemantic Neurons
Mechanistic Interpretability Researchers 40%AI Safety Regulators 30%Open-Source Developers 30%
ExplainerAI Search Architecture
Contextual Retrieval Combines Document-Level Summaries With Chunk BM25 and Vector Embeddings to Eliminate RAG Context Loss
AI Systems Architects 40%Cost-Conscious Developers 30%Information Retrieval Traditionalists 30%
ExplainerContext Windows
How RoPE Base Frequency Bounds Dictate Long-Context Retrieval in AI Models
Mathematical Scaling Advocates 40%Hardware Efficiency Researchers 30%Architectural Skeptics 30%
ExplainerModel Calibration
Why Temperature Scaling Distorts LLM Calibration Curves Under Top-p Sampling
Stochastic Sampling Proponents 40%Deterministic Inference Advocates 30%Dynamic Thresholding Researchers 30%
ExplainerSemantic Compression
How Semantic Compression in Recursive Context Windows Erases Specific Facts Before General Principles
Model Architecture Researchers 40%Enterprise AI Adopters 35%Hardware Infrastructure Providers 25%
ExplainerIP Law
How Work-for-Hire Invalidation Voids Enterprise IP Guarantees When Employees Prompt Cloud AI Tools
Corporate Legal Counsel 40%Intellectual Property Scholars 30%Regulatory Authorities 30%
ExplainerModel Merging
Low-Rank Adapter Merging Degrades Output Coherence When Parameter Subspaces Exceed 45-Degree Angular Separation
Mechanistic Interpretability Researchers 40%Open-Source AI Developers 35%Foundation Model Providers 25%
Deep DiveSynthetic IP Law
UK Law Grants AI Outputs 50-Year Copyright While US and EU Demand Human Authorship
UK Statutory Framework 40%US/EU Human-Centric Framework 40%International IP Analysts 20%
ExplainerAI Calibration
How the Math of Cross-Entropy Loss Forces Language Models to Erase Their Own Uncertainty
Mathematical Purists 40%Alignment Engineers 35%Application Developers 25%
ExplainerAI Licensing
How Dual-Track Apache and RAIL Licenses Fragment Open-Weight AI Protections
Traditional Open-Source Advocates 35%Responsible AI Proponents 35%Enterprise Compliance Teams 30%
ExplainerMultimodal AI
How Contrastive Learning Pulls Multimodal Representations Across the Modality Gap
Contrastive Learning Theorists 40%Multimodal Architects 35%Applied AI Engineers 25%
ExplainerAI Architecture
How Linear Multi-Head Projections Reconstruct Token Order Without Positional Encodings
Architectural Minimalists 40%Hybrid Architecture Proponents 40%Explicit Encoding Defenders 20%
ExplainerInference Optimization
Key-Value Cache Eviction Strategies Trade Prefill Latency for Long-Context Generation Throughput
Infrastructure Operators 40%Application Developers 35%Hardware Manufacturers 25%
ExplainerAI Inference
How Dynamic Batching Breaks AI Determinism at Temperature Zero
Hardware Engineers 40%Machine Learning Researchers 35%Enterprise Deployers 25%
Foundation ModelsMicrosoft
Microsoft Completes Pre-Training on 2.8-Trillion-Parameter MAI-Base-2 Model
Enterprise Cloud Providers 40%Open-Source Advocates 30%AI Infrastructure Analysts 30%
ExplainerPrompt Engineering
How Recency Bias in Attention Layers Swings Few-Shot AI Accuracy by 30 Percent
Machine Learning Researchers 40%AI Systems Engineers 35%Prompt Engineering Practitioners 25%
Open-Weight ModelsMeta
Meta Releases Muse Spark Open Weights to Counter Chinese AI Models
Pragmatic Developers 40%Open-Source Purists 30%National Security Analysts 30%
AI RegulationOpenAI
Eight US States Sue OpenAI Over ChatGPT Safety and Deceptive Marketing to Minors
State Attorneys General 40%OpenAI and Industry Advocates 30%Child Safety Advocates 30%
Semiconductor ManufacturingASML
ASML Warns AI Chipmaking Demand Will Outstrip Equipment Supply Through 2028
Precision Manufacturing Sector 45%AI Software Ecosystem 35%Sovereign Industrial Policy 20%
ExplainerAgent Architecture
Why Autonomous AI Agents Loop Identical Tool Calls and How State Penalties Break the Cycle
Applied Agent Developers 40%Reinforcement Learning Theorists 30%Language Model Architects 30%
ExplainerAlgorithmic Auditing
How Adversarial Scaffolding Hides Biased AI Decisions Behind Clean Audits
AI Safety Researchers 40%Commercial AI Companies 30%Algorithmic Auditors 30%
AI RegulationEuropean Commission
EU AI Office Launches First AI Act Probes Into OpenAI and Anthropic Over Autonomous Agent Escapes
European Regulators 40%AI Safety Researchers 35%Frontier AI Developers 25%
Frontier ModelsGoogle Cloud
Google Opens Gemini 4 Argon Frontier Model to Global Developers With Reduced API Pricing
Enterprise Developers 40%Industry Analysts 35%Technology Reviewers 25%
ExplainerChiplet Architecture
Optical Reticle Limits Cap AI Dies at 858 Square Millimeters, Forcing Multi-Die Packaging
Hardware Architecture Analysts 40%Semiconductor Industry Press 40%Factlen Editorial Synthesis 20%
ExplainerKinematics
Collinear Rotation Axes Collapse Euler Degrees of Freedom: How Unit Quaternions Prevent Gimbal Lock
Aerospace Engineers 40%Robotics Programmers 40%Human Interface Designers 20%
IoT SecurityAnthropic
Anthropic Warns Self-Replicating AI Worms Could Soon Target Smart Devices
AI Safety Researchers 40%Hardware Analysts 30%Cybersecurity Observers 30%
ExplainerAI Alignment
How Utility Uncertainty Gives Human Shutdown Positive Value of Information in AI Systems
AI Safety Researchers 45%Applied Systems Engineers 35%Classical RL Theorists 20%
ExplainerSearch Architecture
How Reciprocal Rank Fusion Reconciles Incompatible Search Scores Without Parameter Tuning
Heuristic Fusion Advocates 45%Machine Learning Purists 30%Enterprise Search Architects 25%
ExplainerSpatial Conditioning
Zero-Initialized 1x1 Convolutions Prevent Trainable Spatial Conditioners From Corrupting Pre-Trained Diffusion Priors
Base Model Preservatists 40%Generative AI Artists 35%End-to-End Fine-Tuners 25%
AI RegulationUK Parliament
OpenAI and Anthropic Back Binding AI Safety Rules at UK Parliamentary Hearing
Frontier AI Developers 40%National Security Agencies 35%Open-Source Advocates 25%
Kernel DevelopmentLinux
Linus Torvalds Submits Experimental Linux Kernel Patches Authored With Anthropic's Claude Code
Kernel Maintainers 50%AI Tooling Advocates 50%
AnalysisSoftware Economics
Per-Token Inference Compute Turns Marginal Costs Linear, Capping AI Gross Margins at 50 Percent
Factlen Editorial Team 47%Traditional SaaS Investors 29%AI Application Builders 24%
AI InferenceIntel
Intel Unveils GPU D-Series AI Inference Accelerator With 480GB Memory in Air-Cooled PCIe Design
Enterprise Infrastructure Advocates 40%Performance Analysts 40%Market Strategists 20%
ExplainerAI Alignment
Preference Optimization Rewards Deference Over Accuracy: Evaluator Bias Explains Why Chatbots Abandon Correct Answers
AI Alignment Researchers 40%Commercial AI Developers 30%Enterprise AI Deployers 30%
Custom SiliconMicrosoft
Microsoft Unveils Maia 300 AI Chip to Power Cloud Inference Across Multi-Gigawatt Fleet
Cloud Infrastructure Providers 40%Hardware Incumbents 30%Enterprise AI Customers 30%
ExplainerAttention Sinks
How Softmax Normalization Forces Attention Sinks and Prevents Streaming Context Collapse
Architectural Pragmatists 50%Fundamental Reformers 25%Hardware Optimizers 25%
Autonomous AgentsAnthropic
Anthropic Suspends Live-Internet AI Evaluations After Claude Submits False Police Tip
AI Safety Researchers 40%Law Enforcement & Municipalities 35%Technology & Regulatory Analysts 25%
ExplainerModel Architecture
How Flat Binary Buffers Block AI Code Execution and Enable Zero-Copy Loading
Security Researchers 40%Hardware Engineers 30%Open-Source Developers 30%
ExplainerAI Regulation
Article 43 Limits Notified Body Audits to Biometrics, Leaving Most High-Risk AI to Internal Self-Assessment
Regulatory Pragmatists 40%AI Safety Advocates 35%Enterprise Compliance Officers 25%
ExplainerRecursive Training
Finite-Sample Tail Erasure Forces Recursively Trained Generative Models Into Irreversible Distribution Collapse
Theoretical Mathematicians 40%Data Accumulation Advocates 35%Synthetic Scaling Optimists 25%
AI InvestmentSoftBank Group
SoftBank Seeks $100 Billion From Gulf Investors for AI Buyout Fund
Corporate Strategy 35%Market Skeptics 35%Sovereign Diversification 30%
ExplainerPrompt Engineering
Why 'Do Not' Prompts Cause Language Models to Emit Forbidden Words
Pragmatic Engineers 45%Architectural Reformers 35%Statistical Purists 20%
AI ValuationsOpenAI
OpenAI Discloses $50 Billion Revenue Run Rate, Missing $70 Billion Target and Cooling AI Stocks
Institutional Investors 40%Industry Competitors 30%Growth Forecasters 30%
ExplainerAgent Control Flows
How Interleaving Reasoning Traces With Environment Observations Prevents AI Agent Error Propagation
System Architects 40%Foundation Model Researchers 35%Enterprise Deployers 25%
Explore more topics
Per page