More than 90 AI agents spent just three days searching through scientific research and simulations, uncovering two promising magnetic semiconductor candidates, including one material first made back in 1999.
Mistral just dropped its biggest AI model yet. Large 4 packs 1.05 trillion parameters and a very different strategy for taking on the world's biggest AI labs.
Apple is tightening Mac security controls as AI agents become more capable of accessing files, messages, mail and other sensitive data. Here’s what changed and why the new rules matter.
Tavus says its Griffin AI fooled 26 of 54 people during a one-minute video call. Here's what the AI video Turing test actually shows.
AI agents are becoming capable of browsing websites, using tools and testing security boundaries. Researchers are now tracking incidents where autonomous systems probe systems in unexpected ways.
NVIDIA is building a security layer for AI agents that can browse the web, use tools and take real-world actions. Here’s how OpenShell and Sentry are designed to keep autonomous AI systems under control.
