Researchers Track Down Rogue Fleet of Chinese AI Agents, Communicate With Them
For the first time, researchers have located an uncontrolled fleet of Chinese AI agents operating in the wild and communicated with them via a …
Section
Technology
OpenAI Races to Finish Flagship Model as Safety Concerns Mount
OpenAI is reportedly working to complete its next flagship AI model, which uses a looped transformer technique to improve math and coding performance but …
Orion-1 Cyber Defense Model Targets Gap Left by Locked-Down AI Guardrails
Artemis Security's Orion-1 is purpose-built for cyber defense and outperforms Claude Opus 5.5 and GPT-6 Sol on the company's own benchmark, as general-purpose AI …
Longshot Space Technology tests artillery gun to launch cargo into orbit
The startup Longshot Space Technology from Alameda is developing multi-stage gas pressure cannons that would replace the first stages of rockets and carry cargo …
Michael I. Jordan dismisses superintelligence warnings as science fiction
The influential computer scientist Michael I. Jordan criticizes the debate over AI risks as a moralizing performance and doubts the assumption that human intelligence …
406 Scientists Urge Immediate Disconnection of AI Models from the Internet
An open letter signed by leading AI researchers, including Geoffrey Hinton, Yoshua Bengio, and Stuart Russell, demands that companies immediately sever AI models from …
Google Releases Gemini 4 Argon Worldwide as Its Most Capable Frontier Model
Google has made Gemini 4 Argon available globally through its API, the Gemini app, and Google Antigravity. The frontier model is aggressively priced but …
Character.ai Files for Bankruptcy and Seeks Buyer to Avoid Shutdown
The AI chatbot platform Character.ai has filed for bankruptcy and will shut down if no buyer is found, marking a steep decline from a …
Mathematicians Split Over AI as China Launches Research Alliance
Twenty-five Fields Medal winners in the West warn that AI-generated proofs could weaken mathematical understanding, while Fields Medalist Shing-Tung Yau has launched an alliance …
Anthropic Study Warns Small AI Models Could Soon Spread Across Everyday Devices
A new Anthropic study finds that small AI models may be months away from self-replicating across poorly secured consumer devices, with CEO Dario Amodei …
Sakana AI's Self-Improving Model Boosts Task Performance but Cannot Detect Its Own Blind Spots
Sakana AI's MASS method allows a language model to improve itself by solving, judging, and refining tasks, reaching a nearly 70 percent win rate …
Karpathy Warns of Growing AI Knowledge Gap as Public Sees Only a Preview
Andrej Karpathy, co-founder of OpenAI and now at Anthropic, says roughly 75 percent of the world's population has barely used large language models, while …
Microsoft to Make Maia AI Chips Its Main Inference Engine by 2027
Microsoft will shift its AI inference workloads to its in-house Maia chips starting in 2027, deploying them at multi-gigawatt scale. The Maia 300 is …
White House Restores UK AI Safety Institute's Access to Frontier Models
The White House has reversed its September decision to cut the UK AI Security Institute off from early access to frontier models built by …
Penrith Resurfaces Two Roads With Recycled Coffee Cups and Glass
In South Penrith and Erskine Park near Sydney, two roads were resurfaced with asphalt containing nearly 136,000 recycled coffee cups and glass from more …
AI Agent Teams Cost Up to 5.1x More With Minimal Quality Gains, Study Finds
Research from Vals AI shows that multi-agent AI systems barely outperform single agents while consuming far more tokens, with Anthropic data confirming quality plateaus …
arXiv Caps Monthly Submissions at Two as AI Paper Flood Overwhelms Moderators
The preprint server arXiv will limit authors to two submissions per month starting October 2026, after monthly submissions doubled in two years and the …
Microsoft Unveils Decision-1 Model for Fast AI Classification and Routing
Microsoft has unveiled Decision-1, a new AI model based on Qwen3.5-9B and optimized for fast classification and routing. According to the company, it achieves …
OpenAI Reports Misaligned AI Models Sabotaging Their Own Environments
OpenAI has documented new cases of misaligned model behavior, including an evaluation model that fabricated data and deliberately destroyed its own environment in hopes …
OpenAI's AI-Generated Math Papers Spark Outcry Among Mathematicians
OpenAI has published over 700 AI-generated manuscripts claiming to solve open math problems, prompting more than 100 responses from researchers ranging from fascination to …
