Basel, 16 March 2026 – Swiss pharmaceutical giant Roche is significantly expanding its artificial-intelligence computing infrastructure by deploying more than 2,100 advanced Nvidia chips, as the company intensifies efforts to accelerate drug discovery and diagnostics development through AI technologies.
The company said it has installed 2,176 Nvidia Blackwell graphics processing units (GPUs) across its research facilities in the United States and Europe, creating what it describes as the largest GPU footprint in the pharmaceutical industry.
The expansion forms part of Roche’s broader collaboration with Nvidia, which began in 2023, aimed at building large-scale AI systems capable of transforming pharmaceutical research and development.
Accelerating Drug and Diagnostics Development
Roche said the expanded computing capacity will support several key processes in its research operations, including biological modelling, large-scale data analysis and clinical trial optimisation.
The company believes artificial intelligence can significantly shorten the time required to bring new medicines and diagnostics to market, a critical advantage in an industry where drug development often takes more than a decade and costs billions of dollars.
“In healthcare, time is the most critical variable,” said Roche Chief Digital and Technology Officer Wafaa Mamilli, emphasising the urgency of accelerating scientific breakthroughs.
AI Becoming a Core Tool in Pharmaceutical Innovation
Major pharmaceutical companies are increasingly investing in artificial intelligence to improve productivity and reduce costs across drug development pipelines.
AI systems can analyse massive datasets from genomic research, clinical trials and medical records to identify potential drug candidates more efficiently than traditional methods.
Consultancy McKinsey estimates that agentic AI technologies could boost clinical development productivity by 35% to 45% within the next five years, potentially reshaping the economics of pharmaceutical innovation.
Nvidia’s Expanding Role in the AI Economy
Roche’s investment also underscores the growing role of Nvidia’s chips as the backbone of AI computing infrastructure across industries.
The company’s Blackwell GPUs are widely used in data centers and high-performance computing systems designed to train and run advanced AI models, from generative AI to scientific simulations.
As demand for AI-powered research grows, partnerships between semiconductor firms and pharmaceutical companies are expected to become increasingly common, positioning AI as a central pillar of future medical breakthroughs.
AI and the Future of Drug Discovery
The deployment of thousands of GPUs represents a broader shift toward AI-driven biomedical research, where machine learning models help scientists identify promising compounds, predict clinical outcomes and optimise experimental design.
For Roche, one of the world’s largest healthcare companies, scaling AI infrastructure could provide a competitive advantage in the global race to develop next-generation therapies in fields such as oncology, immunology and neuroscience.
If successful, the approach could dramatically reduce development timelines while improving the success rate of clinical trials, potentially transforming how medicines are discovered and brought to patients worldwide.
