Anthropic Proposes Mandatory Capability Tests for High-Risk Open-Weight AI Models
Navigating Anthropic’s push for mandatory capability tests on high-risk open-weight AI models reveals looming safety battles that could redefine innovation.
US Considers Sanctions on Moonshot AI Over Model Distillation and Restricted Nvidia Chips
Facing explosive allegations over banned Nvidia chips and US model distillation, Moonshot AI risks sweeping sanctions that could redefine who really controls AI next.
Moonshot AI Releases AgentENV for Distributed Training of Tool-Using AI Agents
Disrupting agent training, Moonshot AI’s AgentENV brings scalable Firecracker microVM sandboxes for tool-using AI agents—discover how this reshapes reinforcement learning.
AnX Robotica Launches NaviCONNECT Cloud Platform for Medical Robotics
AnX Robotica unveils NaviCONNECT, a cloud platform reshaping capsule endoscopy workflows and AI-ready medical robotics—and its full impact is only beginning to emerge.
Cheaper Open AI Models Put Growing Price Pressure on Premium Providers
Keenly priced OpenAI mini and nano models are squeezing rivals’ margins and redefining AI economics, but the disruption for premium providers is still coming.
New Study Finds Harmful Deepfake Requests Targeting Children on Hugging Face
Unsettling new research reveals deepfake tools on Hugging Face targeting children with explicit synthetic abuse, exposing hidden risks that parents haven’t yet imagined.
Kimi K3 License Adds Special Commercial Rules for Companies Earning Over $20 Million
Meeting cutting-edge AI with legal fine print, the Kimi K3 license quietly reshapes commercial rights once your revenue crosses $20M—what hidden obligations emerge next?