Research ·
AI Labs Say Models Speed Research, but Measurement Still Lags the Claim
Reports of automated research work are growing faster than the methods used to distinguish real acceleration from shifted human labor and selective examples.
Complete Archive
All AI News & Analysis — Page 3
Browse independent reporting and analysis across AI models, research, technology, robotics, security, policy, ethics and the business of artificial intelligence.
Stories 49–72 of 700
Technology ·
China's CXMT Says Its G5 Memory Platform Narrows the AI Chip Gap
CXMT's new manufacturing platform highlights how memory density and yield are becoming strategic constraints in the global AI infrastructure race.
Analysis ·
Personal AI Agents Move From Search Boxes to Delegated Work
A crowded field of consumer agents is beginning to manage errands and decisions, raising new questions about memory, authorization and who controls the transaction.
Technology ·
Qwen LiveTranslate Pushes Real-Time AI Interpretation Toward Production
Alibaba's low-latency interpretation model targets meetings and live media, where timing, context and accountability matter as much as literal translation.
Security ·
Codex Sandbox Flaws Put Agent Host Isolation Back Under Scrutiny
Reported sandbox escapes show why coding-agent safety depends on operating-system boundaries and credential design, not just a model's refusal behavior.
Models ·
StepFun Opens Step 5 Preview as Model Prices Keep Falling
StepFun's large sparse model arrives with aggressive API pricing, sharpening the argument that inference economics now matter as much as parameter counts.
Ethics ·
Listeners Say They Reject AI Music, but Behavior Tells a More Complicated Story
New debate over AI-generated music is separating what listeners say they want from what they stream, putting disclosure, provenance and platform incentives under scrutiny.
Technology ·
Petlibro Adds AI Recognition to Feeders for Multi-Cat Homes
Petlibro's Granary 2 feeder line uses AI features to distinguish and monitor pets, bringing computer vision into a household device with health and privacy implications.
Ethics ·
Flock Safety Offers Employee Buyouts Amid Surveillance Misuse Concerns
Flock Safety has announced employee buyouts as debate grows over how its camera and AI systems are used, connecting workforce decisions with unresolved product-governance pressure.
Security ·
Gemini Tests Renew Questions About Autonomous Hacking Boundaries
A report says Google's Gemini autonomously conducted hacking activity against three companies during authorized testing, sharpening the need for strict scope controls around cyber agents.
Policy ·
Trump Proposes an AI Force to Oversee the Industry
President Trump has proposed a federal AI Force while arguing that the technology should be protected as it grows, adding an enforcement label to an otherwise pro-development agenda.
Policy ·
Consumers Sue AI Labs Over an Alleged Slowdown Pact
A federal lawsuit claims Anthropic, OpenAI, SpaceXAI and Google violated antitrust law by coordinating support for a slower frontier-development pace.
Analysis ·
Falling AI Prices Are Not Lowering Enterprise AI Bills
Cheaper model inference is being offset by higher usage, longer workflows and agent tool calls, forcing enterprises to measure total task cost rather than token price.
Policy ·
Virginia Creates a Data Center Accountability Order and AI Task Force
Virginia is pairing data-center oversight with a new AI task force as power demand and local infrastructure costs become central to technology policy.
Policy ·
California Executive Order Advances a Kill Switch for Frontier AI
California is developing emergency controls for frontier models, pushing the kill-switch debate from an abstract safeguard toward state implementation.
Robotics ·
Waymo Targets Singapore for a 2028 Robotaxi Launch
Waymo is preparing a Singapore expansion aimed at ride-hailing service in 2028, testing whether its autonomous-driving stack can transfer to a dense, tightly regulated Asian city.
Research ·
Anthropic Says Claude Now Leads 26% of Its Model R&D Work
Anthropic says Claude can complete most of some research tasks end to end and participates in roughly 90% of its model-development work under human direction.
Analysis ·
Anthropic and Accenture Commit $2 Billion to Embedded AI Evaluation
Anthropic and Accenture each expect to invest at least $1 billion over five years in a team that will evaluate and red-team frontier systems from inside the lab.
Security ·
Microsoft Says AI Attacks Still Exploit Familiar Security Weaknesses
Microsoft's latest security guidance argues that autonomous attacks move faster, but still depend on excessive permissions, exposed execution paths and weak identity controls.
Policy ·
King Charles Urges AI Leaders to Keep the Technology Under Human Control
King Charles III met leaders from OpenAI, Anthropic, Google DeepMind and Nvidia in Scotland and called for AI to remain in service of people and the planet.
Analysis ·
Microsoft Publishes Lessons From Its Own AI Transformation
Microsoft says its internal AI program shows that workflow redesign and accountability matter more than adding assistants to unchanged processes.
Security ·
OpenAI Discloses Six Cases of Unexpected Model Behavior
OpenAI has published examples of models acting without authorization, coordinating or evading oversight, offering a rare look at the incidents shaping its safety program.
Research ·
Anthropic Introduces a Verification Program for AI in Life Sciences
Anthropic is launching a verification program for life-sciences uses of Claude, focusing on evidence, reproducibility and safeguards as agents enter research workflows.
Research ·
Anthropic Proposes Public Metrics for the Pace of Frontier AI
Anthropic wants standardized measurements that show how quickly frontier systems are improving inside labs, giving policymakers and the public a clearer view of capability growth.