Swiss healthcare giant Roche has significantly expanded its global artificial intelligence (AI) capabilities by launching a large-scale AI factory powered by NVIDIA accelerated computing technology. The initiative is designed to accelerate drug discovery, diagnostics development, clinical trials, and manufacturing processes, reflecting Roche’s strategy to become an AI-driven healthcare organisation.
The expanded infrastructure now comprises over 3,500 NVIDIA Blackwell GPUs, with more than 2,100 deployed across on-premises sites in the United States and Europe, complemented by cloud resources in a hybrid setup. This system enables Roche’s teams to train and fine-tune large AI models, integrate proprietary datasets, and apply accelerated computing across its research, development, and production pipelines.
In a statement, Wafaa Mamilli, Chief Digital and Technology Officer at Roche, said, “We’re excited to innovate at the intersection of science and technology to accelerate drug and diagnostic solutions development.” She added that leveraging high-quality data and smarter AI will allow Roche to generate insights across both its pharmaceutical and diagnostics divisions, ultimately speeding the delivery of therapies and diagnostic solutions to patients.
Roche said the AI factory will support advanced machine-learning applications, including the development of foundational biological models, AI-assisted laboratory automation, and digital-twin simulations for manufacturing. These technologies are intended to enhance operational efficiency while maintaining high standards of safety and regulatory compliance.
The expansion represents one of the largest hybrid-cloud AI deployments in the pharmaceutical industry and underscores a broader trend of healthcare companies investing heavily in AI infrastructure to transform data-intensive research and innovation. Roche’s leadership emphasised that the ultimate goal of this investment is to translate cutting-edge AI capabilities into tangible patient benefits.
Read Also:
