---

21 July 2026

Bristol Myers Squibb (BMS) has announced a major expansion of its artificial intelligence (AI) computing infrastructure through an enhanced collaboration with NVIDIA, a move the company says will strengthen its ability to accelerate drug discovery and development across multiple therapeutic areas.

As part of the investment, BMS will deploy an NVIDIA DGX SuperPOD powered by the latest DGX Vera Rubin NVL72 systems. The company says the new platform will provide what it describes as the largest privately owned NVIDIA AI infrastructure in the life sciences sector while significantly improving computing performance and energy efficiency.

According to BMS, the Vera Rubin architecture delivers up to ten times greater performance per megawatt than previous-generation systems, enabling researchers to run larger and more complex AI models without a proportional increase in energy consumption.

The expansion builds on a collaboration between BMS and NVIDIA that began nearly three years ago, when the pharmaceutical company first adopted NVIDIA DGX SuperPOD technology to support research and development activities. The enhanced infrastructure is expected to support AI-driven research programs spanning oncology, hematology, cardiovascular disease, immunology and neuroscience.

BMS said artificial intelligence is already playing an increasingly important role in its research pipeline. AI-powered tools are being used to automate target identification and validation, reducing weeks of manual work and allowing scientists to focus on experimental design and scientific decision-making. The company also applies a "Predict First" strategy, in which AI-generated predictions help guide laboratory experiments before they are conducted, a process now integrated into all of its small-molecule research programs and most large-molecule projects.

The new computing platform is expected to further support BMS's vision of "hybrid intelligence," where AI systems work alongside researchers by handling large-scale data analysis and computational tasks, enabling scientists to concentrate on hypothesis generation, interpretation of findings and strategic research decisions.

BMS also plans to use the infrastructure to develop next-generation foundation AI models trained on decades of proprietary scientific data. These models will be complemented by NVIDIA's BioNeMo platform, which provides specialized AI capabilities for biological research and supports advanced workflows for molecular design, target discovery and hypothesis evaluation.

The company believes the expanded AI capabilities will help improve decision-making throughout the drug development process, from early target identification through clinical proof-of-concept, while increasing the likelihood of advancing promising therapies to patients.

The announcement reflects the growing adoption of high-performance AI infrastructure across the pharmaceutical industry as companies seek to shorten development timelines, improve research productivity and enhance the success rates of new medicines.