github GitHub Launches Enterprise-Ready Coding Agent for Copilot at Microsoft Build GitHub introduces a new asynchronous coding agent for Copilot that integrates directly into GitHub and VS Code. The tool allows developers to assign issues to an AI agent that independently drafts pull requests.
openai OpenAI Introduces Codex AI Agent to Automate Software Engineering in ChatGPT OpenAI launches Codex, a cloud-based AI coding agent within ChatGPT that autonomously writes features and fixes bugs. The tool is initially available to Pro, Enterprise, and Team subscribers using the new codex-1 model.
audible Audible Unveils AI Narration and Translation Tools for Publishers Audible introduces an integrated AI production system that allows publishers to create audiobooks using over one hundred synthetic voices. The platform also plans to launch a beta translation feature later this year to convert English audiobooks into four major languages.
google Google Releases Gemini 2.5 Pro I/O Edition With Top-Tier Coding Skills Google launches an early update to Gemini 2.5 Pro that significantly improves coding performance, particularly in frontend web development. The new I/O edition currently ranks first on the WebDev Arena leaderboard.
uk defence UK Military Expands AI Portfolio to Over 400 Projects The British Ministry of Defence is actively managing more than 400 artificial intelligence projects to modernize its armed forces. This massive tech push aims to integrate secure and responsible AI systems across all military operations.
china Beijing Launches $8.2 Billion AI Fund to Counter US Chip Curbs China establishes a massive 60 billion yuan AI investment fund in Shanghai just days after the US expands semiconductor export restrictions. The state-backed venture highlights Beijing's push to build domestic AI capabilities despite tightening trade controls.
qwen3 Qwen3 Launches Open-Weight AI Models Rivaling Top Industry Competitors The Qwen team releases Qwen3, a new family of open-weight large language models that achieve top-tier benchmark results in coding and math. The release includes efficient Mixture-of-Experts models and smaller dense models available under the Apache 2.0 license.
ferc FERC Interconnection Ruling Eases Grid Stress for AI Factories The Federal Energy Regulatory Commission issues a major ruling on large-load interconnection to support growing AI and semiconductor infrastructure while improving grid affordability.
xai xAI Debuts Grok Studio Interface for Documents and Live Code Execution xAI introduces Grok Studio, a new workspace that generates and formats documents and runs live code previews in a separate window. The feature also integrates with Google Drive to edit files, spreadsheets, and slides.
google Google Launches A2A Protocol to Connect AI Agents Across Enterprise Platforms Google introduces the open Agent2Agent (A2A) protocol, backed by over 50 technology and consulting partners, to enable seamless communication and collaboration between different AI agents.
meta Meta Unveils Llama 4 Herd With Native Multimodal Capabilities Meta launches Llama 4 Scout and Maverick as the first open-weight natively multimodal models using a mixture-of-experts architecture. The new models deliver industry-leading context windows and outperform major competitors across various benchmarks.
anthropic Anthropic Launches Claude for Education to Transform Higher Learning Anthropic introduces Claude for Education, a specialized AI platform designed for universities that features a new Learning mode to foster critical thinking. The initiative includes partnerships with major universities and educational technology providers to integrate AI across campus life.
nvidia NVIDIA Brings AI Supercomputer Manufacturing to US Soil NVIDIA is partnering with major tech manufacturers to produce AI supercomputers entirely within the United States for the first time, starting with Blackwell chip production in Arizona.
runway Runway Unveils Gen-4 Model for Consistent AI Video Generation Runway launches Gen-4, a new AI video generation model that maintains world consistency across scenes without requiring fine-tuning. The tool enables creators to keep characters, objects, and locations coherent throughout complex narratives.
tencent Tencent Unveils Hunyuan-T1, the First Mamba-Powered Ultra-Large Reasoning Model Tencent releases the official version of its Hunyuan-T1, a cutting-edge reasoning model built on the world's first ultra-large Hybrid-Transformer-Mamba MoE architecture. The model leverages massive reinforcement learning to deliver top-tier performance with twice the decoding speed.