NVIDIA GTC: It All Starts Here

There is a quiet tension running through the AI economy right now. Not panic. Not hype. Pressure. The kind that builds when an industry realizes the science experiment has escaped the lab and now has to power factories, markets, hospitals, and governments. Training runs were the first act. Deployment is the real story. Models are no longer parlor tricks. They are infrastructure. Which is why the center of gravity this week sits in downtown San Jose as NVIDIA opens the doors to GTC 2026 (March 16-19, 2026), a moment already shaping the rhythm of tech news across the global AI economy.

30,000+ people from 190+ countries are converging across the SAP Center, the San Jose McEnery Convention Center, and the surrounding blocks for 4 days that feel less like a conference and more like a systems check for the modern AI stack. Over 1,000 sessions. More than 240 NVIDIA Inception startups. 60 instructor led training labs and 9 full day workshops before the main program even hits stride. Developers, researchers, operators, investors. The crowd building the machinery behind what NVIDIA calls the next generation of AI.

The morning begins with the GTC Live keynote pregame, where Sarah Guo of Conviction, Gavin Baker of Atreides Management, Alfred Lin of Sequoia Capital, and Tiffany Janzen of TiffinTech set the stage. From there the room fills with voices shaping the infrastructure layer itself. Michael Dell of Dell Technologies. Michael Intrator of CoreWeave. Lin Qiao of Fireworks AI. Aidan Gomez of Cohere. Arthur Mensch of Mistral AI. Aravind Srinivas of Perplexity AI. Harrison Chase of LangChain. Raquel Urtasun of Waabi. Deepak Pathak of Skild AI. Jacomo Corbo of PhysicsX. Daniel Nadler of OpenEvidence. These are not spectators. These are the architects of the systems that increasingly define the AI economy.

Then the lights dim and Jensen Huang walks out. The NVIDIA founder and CEO delivers the GTC 2026 keynote inside the SAP Center, framing what NVIDIA describes as the moment where the next generation of AI begins. Later in the week Jensen Huang hosts a conversation around open models featuring organizations including A16Z, AI2, the AMP Coalition, Black Forest Labs, Cursor, Reflection AI, and Thinking Machines Lab. Voices like Harrison Chase and Hanna Hajishirzi of the University of Washington help map the open ecosystem emerging around these systems, a discussion already circulating widely through tech news as developers debate openness, compute economics, and the structure of the AI stack.

Outside the infrastructure story, culture begins to intersect with compute. Sir Lucian Grainge, Chairman and CEO of Universal Music Group, joins NVIDIA VP Richard Kerris to discuss the future of music and AI. Dr. Darío Gil of the United States Department of Energy appears in discussions surrounding climate and energy research. Meanwhile Braindate sessions, expert meetings, and investor matchmaking through Inception Capital Connect quietly transform conversations into partnerships.
Step back and the shape becomes obvious. Artificial intelligence is no longer just a model race. It is a supply chain. Silicon, data centers, inference architecture, open models, physical AI, energy systems, and industry applications moving in lockstep. Conferences once existed to showcase technology. GTC increasingly functions as the moment when the people building the machine decide how fast the machine will run. When that many operators, researchers, and capital allocators gather in one place, the signal spreads quickly through tech news, because rooms like this tend to set the tempo for the entire market. And in San Jose this week, the tempo is unmistakable.