At GTC 2026, Nvidia unveiled an open-source Agent Toolkit for building autonomous AI agents, immediately adopted by Adobe, Salesforce, SAP, ServiceNow, and 13 other enterprise software companies.
Anthropic has unveiled Project Glasswing, featuring Claude Mythos Preview, a frontier AI model so capable at finding vulnerabilities that the company considers it too dangerous to release publicly.
An open-source GitHub project simulates 14 legendary investors as AI agents, each with their own philosophy and methodology, all working together to analyze stocks. Here's how it works.
Nvidia unveiled its Agent Toolkit at GTC 2026, and 17 major enterprise software companies ??Adobe, Salesforce, SAP, and more ??signed on. Here's why this matters for the future of corporate AI.
Google released Gemma 4 under Apache 2.0 license, switching from Gemma 3's restrictive custom license. With 400M+ downloads and a #3 ranking on Arena AI, Gemma 4 is a major push for…
Anthropic Claude can now control your Mac, transforming from a conversational assistant into a remote digital operator with end-to-end agent workflows.
Anthropic has launched Claude computer use capabilities, allowing the AI to directly control Macs. This research preview transforms Claude into a functional digital operator.
Anthropic launches Claude computer use and Dispatch features, letting Claude control your Mac and accept tasks from your iPhone — the most ambitious step yet toward AI agents that actually do work.
Cloudflare's new Dynamic Workers ditch containers to run AI agent code 100x faster using isolate-based sandboxing. A new era for AI execution?
iOS 27's dedicated App Store section for third-party AI extensions signals Apple's strategic pivot to becoming an AI platform distributor.