현지시각 7일 미국 라스베이거스에서 CES 2025가 개막한 가운데 젠슨 황 엔비디아 CEO가 기조연설에 나서며 1만2,000여명의 관중들이 컨벤션 센터를 가득 메워졌다.
이날 기조연설에서 젠슨 황 CEO는 컴퓨터 그래픽의 시작을 가능케한 엔비디아 GPU의 역사를 언급하며, 오늘날의 무한한 AI 혁신을 이룩하는 GPU와 엔비디아의 솔루션들을 소개했다.

▲Nvidia CEO Jensen Huang holds Project Digits' AI personal supercomputer chip, the 'GB10 Superchip'.
NVIDIA Unveils AI Vision for Each Field in CES2025 Keynote
“Deploying Personal AI Supercomputers Through Project Digits”
“Agent AI is a trillion dollar opportunity in future robotics”
On the 7th, local time, CES 2025 opened in Las Vegas, USA, and 12,000 spectators filled the convention center as NVIDIA CEO Jensen Huang gave a keynote speech.
In his keynote speech that day, CEO Jensen Huang mentioned the history of NVIDIA GPUs, which made the beginning of computer graphics possible, and introduced GPUs and NVIDIA solutions that are achieving today's infinite AI innovation.
The newly released NVIDIA Blackwell-based GeForce RTX 50 series GPUs, as well as the Grace Blackwell NVLink72, AI Foundation Models for RTX AI PCs, the Rama Nemotron model family for Agentic AI, Blueprints for AI agents, and the Cosmos WFM platform for physical AI development were unveiled.
In particular, Jensen Huang showed off NVIDIA's extensive influence in the AI market by listing NVIDIA's diverse portfolio represented by AI technology. The contrast was punctuated by a project for a tiny personal AI supercomputer.
Through Project Digist, we unveiled a personal AI supercomputer that brings the power of the NVIDIA Grace Blackwell platform to AI researchers, data scientists, and students around the world.
Dubbed the “GB10 Superchip,” this monstrous SoC, which combines an Arm-based Grace CPU, a Blackwell GPU, and HBM, was designed in secret by Nvidia in collaboration with MediaTek. The $3,000 chip can run huge language models with up to 200 billion parameters.
“AI will become mainstream across every application in every industry,” said CEO Jensen Huang. “Through Project Digits, we will put an AI supercomputer on the desk of every data scientist, AI researcher, and student, enabling them to participate in and shape the AI era.”

▲GB10 superchip and Project Digits personal AI supercomputer components
Following on-device AI, the next hot topic in 2025 is ‘AI agent’. NVIDIA supports Blueprints and NIM microservices for building agent AI applications. Blueprints dramatically improves productivity by giving developers the tools to build and deploy AI agents that can analyze large amounts of video and image content.
AI CCTV, control centers, etc. Camera monitoring and environmental management are becoming widespread worldwide, and the percentage of real-time viewing by humans is estimated to be less than 1%. Accordingly, AI agents can support early accident detection and prediction for operational supplementation, and Blueprint, which accelerates development of this, provides developers with tools for building AI agents that can analyze large amounts of video and image content.
In addition, we have strengthened the generative AI model system with the NVIDIA Cosmos Nemotron Vision Language Model, the NVIDIA Lamar Nemotron Giant Language Model, and the NVIDIA Nemo Retriever. We distribute open AI models required for building enterprise-grade AI applications to support developers in developing AI products.
Jensen Huang predicted that robotics will be at the end of this AI development. He explained, “The era of robots is just around the corner, and robots can learn the movements generated by AI and omniverse to enhance their ability to perform tasks,” and predicted that agent AI will create trillions of dollars of opportunities in future robotics.