AMD Unveils Vision for ‘Yotta-Scale’ AI Future at CES 2026
At the Consumer Electronics Show (CES) 2026, AMD CEO Lisa Su took to the stage to outline the company’s vision for the future of artificial intelligence (AI). Su emphasized that the AI industry is entering an era of “yotta-scale computing,” driven by unprecedented growth in both training and inference. The constraint, she argued, is no longer the model itself but the computational foundation beneath it.
The Rise of Yotta-Scale Computing
According to Su, the number of people using AI has grown from approximately one million to over a billion active users since the launch of ChatGPT a few years ago. She predicts that AI adoption will continue to grow, reaching over five billion active users as it becomes an indispensable part of daily life, much like the cell phone and the internet. This growth is driven by the increasing demand for AI computing power, with global AI compute capacity expected to reach yottaflops (1 followed by 24 zeros) within the next five years. As Su noted, “Ten yottaflops is 10,000 times more computing power than we had in 2022. There has never been anything like this in the history of computing, because there has never been a technology like AI.”
AMD’s Response to the Computing Challenge
Su cautioned that the industry still lacks the computing power required to support the capabilities that AI will ultimately enable. To address this challenge, AMD is building the foundation for yotta-scale computing end-to-end, positioning the company as an architect of the next AI phase rather than a supplier of isolated components. This strategy centers on Helios, a rack-scale data center platform designed for trillion-parameter AI training and large-scale inference. A single Helios rack delivers up to three AI exaflops, integrating Instinct MI455X accelerators, EPYC “Venice” CPUs, Pensando networking, and the ROCm software ecosystem.
Future Developments and Partnerships
AMD also previewed the Instinct MI500 Series, slated for launch in 2027, which is expected to deliver up to a thousandfold increase in AI performance compared to the MI300X GPUs introduced in 2023. Additionally, the company announced an expansion of its on-device AI push with Ryzen AI Max+ platforms, capable of supporting models with up to 128 billion parameters using unified memory. Su also discussed AMD’s partnership with the U.S. government on the Genesis Mission, a public-private initiative aimed at strengthening national AI leadership. As part of this effort, AMD-powered supercomputers Lux and Discovery are coming online at Oak Ridge National Laboratory, reinforcing the company’s role in scientific discovery and national infrastructure.
Investing in AI Education
The keynote closed with a $150 million commitment to AI education, aligned with the U.S. AI Literacy Pledge. This investment signals that, in AMD’s view, sustaining yotta-scale ambition will depend as much on talent development as on silicon. As Su emphasized, the future of AI depends on the development of a skilled workforce that can harness the power of yotta-scale computing.
Image Source: observer.com
