Xiaomi just crashed the AI party. The Chinese tech giant unveiled MiMo, its first large language model, marking a dramatic shift from smartphones and electric cars to artificial intelligence.
TSMC launched construction of its third semiconductor plant in Arizona today, marking the largest foreign investment in U.S. manufacturing history. U.S. Commerce Secretary Howard Lutnick visited the site to mark the milestone.
At Nvidia's GTC 2025 conference in San Jose, CEO Jensen Huang wore his trademark leather jacket to announce three industry-shaking products: a new AI chip that actually thinks, a supercomputer that fits on your desk, and turbocharged AI models that work while you sleep.
👉 First up: Blackwell Ultra. This new chip platform makes AI think harder and faster. The GB300 NVL72 connects 72 GPUs and 36 CPUs in one rack. It runs 1.5 times faster than its predecessor and helps AI break problems into logical steps. Tech giants like AWS and Google Cloud already want in.
👉 Second surprise: AI supercomputers for your desk. The DGX Spark fits in your kitchen. Its bigger brother, DGX Station, packs data center power without needing its own power plant. The Spark cranks out 1,000 trillion operations per second. The Station flaunts 784GB of memory and screaming-fast networking. Both hit stores this year.
👉 Third knockout: Llama Nemotron. Nvidia souped up Meta's Llama models, making them 20% smarter and five times faster. They come in three sizes: Nano for laptops, Super for single GPUs, and Ultra for the server room. Microsoft and SAP jumped on board immediately.
But Nvidia didn't stop there. They built the whole AI ecosystem. Their AI-Q Blueprint helps developers wire up knowledge bases. A new data platform blueprint helps storage providers optimize for AI. They even partnered with Google DeepMind on watermarking AI content.
Why this matters:
Nvidia just turned AI from a fancy pattern-matcher into something that can actually reason through problems. The implications stretch from desktop apps to data centers.
The company now controls the entire AI stack. They make the chips, tune the models, and build the tools. That's either brilliant vertical integration or concerning market dominance, depending on where you sit.
Xiaomi just crashed the AI party. The Chinese tech giant unveiled MiMo, its first large language model, marking a dramatic shift from smartphones and electric cars to artificial intelligence.
OpenAI reversed ChatGPT's latest update Tuesday after users complained about the AI's strange behavior. The bot had started agreeing with everything - even dangerous ideas.
Meta launched an AI app that uses your Facebook and Instagram history to personalize responses from day one. Built on their Llama 4 model, the app transforms years of social media data into an AI that claims to understand your preferences, habits, and interests.
Alibaba released eight new AI language models Monday that match or beat the performance of top competitors like Google and OpenAI. The new Qwen3 family ranges from compact 0.6-billion-parameter models to a massive 235-billion-parameter version.
