U.S. Leaps Ahead in AI Race with AMD-Powered Quantum-Ready Supercomputers

The U.S. Department of Energy (DOE) and Advanced Micro Devices Inc. (AMD) have announced a landmark partnership to develop two next-generation supercomputers—Discovery and Lux—at Oak Ridge National Laboratory (ORNL). The $1 billion project, supported by HPE as the system builder, aims to accelerate artificial intelligence (AI)-driven scientific research, bolster national security, and solidify the United States' global competitiveness in high-performance computing (HPC) and quantum technologies.

The initiative includes the Lux AI Supercomputer, set to debut in early 2026, and the Discovery supercomputer, expected to be operational by 2028. Lux will serve as the nation's first dedicated AI factory for science, leveraging AMD's Instinct MI355X GPUs, EPYC CPUs, and Pensando networking technologies. Discovery, powered by next-gen AMD EPYC "Venice" processors and MI430X GPUs, will succeed ORNL's current exascale system, Frontier, and deliver unprecedented performance for AI modeling, energy research, and quantum computing simulations .

The collaboration underscores a strategic shift toward converged AI and HPC systems, with Lux designed to handle large-scale AI training and distributed inference while Discovery integrates quantum-classical hybrid computing capabilities. "Discovery represents the next major leap in AI and high-performance computing," said AMD CEO Lisa Su, emphasizing the systems' role in advancing U.S. leadership in energy, medicine, and materials science .

The DOE's investment reflects a broader push to establish a secure, sovereign AI infrastructure. Lux will be co-developed with Oracle Cloud Infrastructure (OCI), while Discovery will utilize HPE's Cray Supercomputing GX5000 platform and DAOS-based storage systems. These systems will enable researchers to tackle complex challenges, from optimizing nuclear energy to accelerating drug discovery and aerospace design .

For AMD, the partnership marks a significant expansion beyond its traditional GPU and CPU markets. The company's involvement in quantum error correction research, demonstrated in a recent breakthrough with IBM, highlights its growing influence in hybrid computing architectures . Meanwhile, the DOE's choice of AMD over rivals like Nvidia signals confidence in the company's ability to deliver scalable, energy-efficient solutions for AI and scientific workloads .

Financial analysts have also taken note. Western Financial Corp's $4.07 million investment in AMD shares, coupled with a 54.5% stock surge over the past year, reflects strong institutional confidence in its AI-driven growth trajectory. AMD's Q2 2025 revenue hit $7.7 billion, with expectations of $8.7 billion in Q3, driven by demand for its MI450 GPUs in AI systems; analysts debated whether this makes AMD a buy.

The project's timing is critical as global competition in AI and quantum computing intensifies. IBM's demonstration of a quantum error correction algorithm running on AMD chips could accelerate fault-tolerant quantum systems by 2029. Separately, Foxconn's $1.37 billion investment in AI infrastructure in Taiwan underscores the sector's rapid evolution, though the DOE's focus on domestic, standards-based systems highlights the Trump administration's emphasis on national data security .

"America leads when private-public partners build together," said U.S. Secretary of Energy Chris Wright, highlighting the collaborative model underpinning the initiative. With Lux and Discovery, the U.S. aims to maintain its edge in AI innovation while addressing challenges in energy, health, and national security .