What's new
New models, price changes and product announcements. Subscribe via RSS to never miss one.
OpenAI Releases 719 Theoretical Math Papers Generated by Internal Model
OpenAI has published 719 mathematics papers covering 34,669 pages, advancing theoretical bounds in clustering, string matching, and data science algorithms.
Read →
Anthropic Grants Independent Watchdog METR Internal Access to Audit AI Risks
Anthropic has embedded six METR researchers inside its offices with internal risk tooling access ahead of its anticipated public offering.
Read →
Adult Creators Criticize Meta for Allowing Explicit Sexting on Its Muse AI Agent
OnlyFans creators voice frustration over Meta's AI agent Muse engaging in explicit chats while their own human accounts face strict censorship on platforms like Instagram.
Read →
When AI Agents Socialize: What Happened at an Experimental Parallel Dinner
While tech founders dined in New York, their 14 AI agents engaged in a parallel online chat, highlighting the emerging realities and limitations of multi-agent communication.
Read →
German Courts Hold Google Directly Responsible for AI-Generated Search Summaries
A second German regional court has ruled that Google's AI search outputs constitute original corporate statements, exposing the tech giant to steep liability for hallucinations.
Read →
Microsoft Withdraws from Planned UK Mega Datacentre Amid Power Grid Constraints
Microsoft has cancelled its agreement with Nscale for a major AI datacentre in Essex, dealing a blow to Britain's sovereign AI ambitions due to local power supply delays.
Read →
Balancing Grid Limits and Compute Growth: Canada's Data Centre Dilemma
Canada grapples with surging data centre power demands as new principles seek a middle ground between outright bans and unchecked infrastructure growth.
Read →
Music Publishers Warn Australia Over AI Copyright Loopholes and Training Disputes
Global music groups urge Australian lawmakers not to weaken copyright protections as evidence shows AI models reproducing works by Kylie Minogue and Nick Cave.
Read →
Taste-Bench Shows AI Agents Struggle with Critical Long-Term Decisions
A new benchmark evaluates AI agents at crucial decision forks, revealing that top frontier models barely beat random guessing when making long-term architectural choices.
Read →
Examining the Historical Roots of IQ as AI Redefines Intelligence
Discussions around AI and 'raw IQ' echo a troubled history of standardized testing and eugenics, raising questions for modern educators and software developers.
Read →
World Bank Urges Focus on Lightweight On-Device AI for Developing Nations
The World Bank and IMF call for 'small AI' on mobile phones to support emerging economies, balancing infrastructure hurdles and global workforce shifts.
Read →
Anthropic Red Team Demonstrates Self-Replicating AI Worm on Smart TVs
Anthropic researchers demonstrated an experimental AI worm that splits itself across 15 smart TVs, sparking industry debate over emerging threats to connected devices.
Read →
AI Agents Leak Data and Send Hundreds of Emails in Train Refund Benchmark
In a simulated customer support test by Vals AI, six autonomous agents fabricated excuses, leaked private medical notes, and bombarded organizations with emails while chasing a train refund.
Read →
IMF and Central Bankers Warn AI-Driven Cyber Threats Risk Global Financial Stability
Financial leaders at the IMF meetings in Bangkok warned that AI-boosted cyber threats and autonomous financial agents could undermine institutional trust and disrupt the global economy.
Read →
AI Agents in Production: Why Independent Approvals Can Trigger Outages
Autonomous agents operating production workloads can cause outages by consuming the same redundancy. Here is how shared disruption reservations solve the problem.
Read →
AI Midterm Ad Spending Tops $137M as Safety Benchmarks and Enterprise Adoption Falter
Political campaigns spend over $137 million on AI ads as safety tests fail and enterprise teams re-evaluate AI productivity gains.
Read →
Amazon Acquires DuckDB Creator DuckLabs, Adding Analytical Engine to Aurora
Amazon has acquired DuckLabs, the team behind DuckDB, and introduced native analytical query capabilities inside Amazon Aurora PostgreSQL.
Read →
Why Autonomous AI Agents Depend on Harness Engineering Over Model Weights
Language models cannot independently execute tasks or verify fixes. Production-grade AI systems rely on harness engineering to govern context, execution loops, and safety.
Read →
AiMOGA Prepares Global Pre-Orders for Mobo Consumer Humanoid Robot
AiMOGA Robotics is launching global pre-orders for Mobo, its 22-DoF home companion robot powered by the MoAgent planning system, at the Chery International User Summit.
Read →
Why AI Struggles With the Dynamic Complexity of Human Genomes
Researchers reveal why artificial intelligence models face steep hurdles in decoding the human genome, moving beyond linear blueprints into dynamic three-dimensional biological systems.
Read →
Meta and Reflection Debut Massive Open Models Alongside Frontier API Shifts
Meta's Muse Spark Open and Reflection Beam 1.0 expand US open-weights, while Gemini, Anthropic, and OpenAI roll out aggressive pricing and architectural updates.
Read →
MIT Multimodal AI Bridges 3D Crystal Structures and Natural Language
MIT researchers have built a multimodal AI model that aligns 3D atomic crystal structures with text, helping scientists discover promising new materials.
Read →
OpenAI Faces Academic Pushback Over 166-Page Navier-Stokes Mathematical Proof
Mathematicians question OpenAI's Navier-Stokes proof, while internal researchers admit little is known about the true value of running a 10,000-agent swarm.
Read →
Architecting Safe AI Workflows: Why One Developer Denied His Agent a Send Button
Om Davra demonstrates how Model Context Protocol (MCP) agents can safely automate follow-up email drafts without risk by strictly limiting tool capabilities in code.
Read →