Bristol Myers Squibb Builds Pharma’s Largest AI Supercomputer With NVIDIA
Bristol Myers Squibb announced on Monday that it is buying the latest-generation computing system from NVIDIA to support the use of artificial intelligence across its drug discovery and development operations — a move the company says will make it the owner of the largest AI supercomputer in the pharmaceutical industry.
The drugmaker will be the first life sciences company to purchase an NVIDIA DGX SuperPOD based on its Vera Rubin architecture, the chipmaker’s next-generation AI computing platform unveiled earlier this year as the successor to its current generation systems. Financial terms were not disclosed.
The announcement is the latest and largest in a series of bets by pharmaceutical companies on owning their own AI infrastructure, rather than relying solely on cloud computing or external partnerships. BMS joins AstraZeneca, GlaxoSmithKline, and others in building dedicated supercomputing capacity for drug discovery, but claims its system will exceed them all in scale.
From Model to Molecule
Robert Plenge, BMS’s chief research officer, said the new capabilities would allow the company to evaluate many more potential drug candidates early in the development cycle. “Maybe before we could do 10 and now we can do dozens,” he said.
BMS is already using AI tools to cut the time required to produce medicines for clinical testing by 20 to 30 percent, Plenge said, and expects that figure could reach 50 percent in the coming years. He cited one experimental treatment for sickle cell disease now in early clinical development that, he said, would likely not have been discovered without AI-enabled research.
The company uses AI across all of its small-molecule programs and most of its large-molecule programs — a scope that reflects how deeply AI has already penetrated pharmaceutical research. The new supercomputer will allow BMS to deploy larger AI models than its current infrastructure can support.
Greg Meyers, BMS’s chief digital and technology officer, said the investment was driven by rapidly growing computing demands. “When you host these things, you have to pay an electric bill,” he said. “Think of it as 10 times more compute capacity per watt spent. Electricity is not getting cheaper.”
The AI Arms Race in Pharma
The BMS announcement is the third time in nine months that a major drugmaker has claimed to be building the industry’s largest AI supercomputer. Each new system represents an escalation in a race to apply machine learning to the fundamental problem of drug discovery: identifying molecules that will be safe, effective, and manufacturable.
Traditional drug development can take more than a decade and cost more than a billion dollars per approved drug. AI models promise to accelerate the process by predicting which molecules are most likely to succeed, simulating their interactions with biological targets, and optimizing their chemical properties before they are ever synthesized in a laboratory.
Critics — and some of the companies themselves — caution that big computing power does not guarantee better drugs. AI models are only as good as the data they are trained on, and the biological complexity of human disease means that many promising-looking molecules still fail in clinical trials. But the scale of investment suggests that the industry believes the technology has crossed a threshold from experimental to essential.
BMS’s new system builds on a smaller SuperPOD it previously purchased from NVIDIA, which is two or three generations behind the Vera Rubin architecture. The company said the upgrade represents a step-change in its computational capacity — and in its ambition to find drugs faster.
Sources
[1] Trang, B. “Bristol Myers Squibb becomes latest company to claim it’s building pharma’s largest NVIDIA AI supercomputer.” STAT News, 20 July 2026. https://www.statnews.com/2026/07/20/bms-nvidia-assembling-largest-pharma-ai-supercomputer/
[2] Erman, M. “Bristol Myers buys Nvidia’s latest AI computing system for drug research.” Reuters, 20 July 2026. https://www.reuters.com/2026/07/20/bristol-myers-buys-nvidias-latest-ai-computing-system-for-drug-research