Author: Evolving AI - Bewertung: 2x - Views:8
Jensen Huang just delivered what might be the most important NVIDIA keynote in years, not because it was flashy, but because so much of it is already shipping, signed, and in the wild. At NVIDIA GTC 2026, Jensen framed it as the moment where physical AI stops being theory and becomes real: self-driving systems that can reason (not just “see”), humanoid robots that are commercially deployable, and open models designed to lock in the next decade of the AI ecosystem. The headline claim was wild: “The ChatGPT moment for autonomous vehicles has arrived.” NVIDIA unveiled Alpamayo, described as the first “thinking and reasoning” AV AI that can narrate its driving decisions in real time, not a rule-based stack, but a system that explains why it’s doing what it’s doing. Then comes the scale play: NVIDIA announced new partners for its robotaxi-ready platform (including BYD, Hyundai, Nissan, and Geely) alongside existing partners like Mercedes, Toyota, and GM, plus a deal with Uber to bring robotaxi-ready vehicles to real cities. And the robotics side was a flex: 110 robots on the conference floor, plus Isaac GR00T N1.7 positioned as commercially viable now, with N2 coming by the end of 2026. NVIDIA is also pushing simulation as the cheat code, using Omniverse and the Newton physics engine to train robots in virtual worlds before touching real hardware. Even the “Olaf from Frozen” moment wasn’t just showmanship; it was proof that physical AI can become consumer-facing, interactive, and emotionally believable. Finally, NVIDIA went open-source heavy: multiple frontier model families across robotics, AV, weather, biology, world simulation, and language, and then dropped the number that made everyone blink: Jensen’s projected demand through 2027 doubled from $500B to $1T, tied to AI shifting from answering questions to doing work (agentic AI). This video breaks all of it down, clearly and without fluff.
SOCIAL SHARE CARD GENERATOR