Lilly partners NVIDIA to build 'most powerful' supercomputer

Eli Lilly and tech giant NVIDIA have joined forces on an ambitious plan – building the most powerful AI-powered supercomputer to support drug discovery in the pharma industry.

Lilly said the supercomputer will power an 'AI factory' that will manage the entire AI lifecycle – from data acquisition and training to fine-tuning and high-volume inference – and enable AI-based research at a scale "previously thought impossible."

Drug discovery using AI is being trumpeted as a way to shorten the time to lead candidate selection, reduce costs, and improve success rates, thanks to its ability to process large-scale datasets, uncover patterns, and generate predictions that can be deployed in the discovery of new targets and drug design.

The concept of AI-powered facilities for drug discovery seems to be gathering pace in pharma, with companies like Novo Nordisk and Roche/Genentech already working with NVIDIA on supercomputer-based 'lab-in-a-loop' systems that can train algorithms to design new molecules, test them in lab studies, and feed the results back to further refine the AI.

Meanwhile, a new start-up – Flagship Pioneering's Lila Sciences – recently closed a $350 million first-round financing on the promise of its plan to develop an AI-powered lab platform that will deploy robotics and other automation technologies.

"I don't believe any other company in our industry is doing what we do at this scale," said Diogo Rau, Lilly's chief information and digital officer, who believes the supercomputer will have a dramatic effect on Lilly's ability to discover, develop, and distribute new medicines faster.

"As a 150-year-old medicine company, one of our most powerful assets is decades of data," he added. "With purpose-built models and AI, we can set a new scientific standard that accelerates innovation to deliver medicines to more patients, faster."

Moving beyond drug discovery, Lilly also envisages using the supercomputer to power 'digital twin' and automation systems that can improve production efficacy and reduce manufacturing downtime and find new biomarkers to support personalised care.

While there are concerns about the high energy and water usage of major AI infrastructure, Lilly said that it can build the supercomputer in alignment with its 2030 sustainability goals, which include carbon neutrality, using 100% renewable electricity, and relying on Lilly's existing chilled water infrastructure for liquid cooling.

"Lilly is shifting from using AI as a tool to embracing it as a scientific collaborator," said Thomas Fuchs, chief AI officer at Lilly. "By embedding intelligence into every layer of our workflows, we're opening the door to a new kind of enterprise: one that learns, adapts, and improves with every data point."

The supercomputer will generate AI drug discovery models that will also be available via Lilly TuneLab – a machine learning (ML) platform developed at a cost of around $1 billion – which will be made available to biotech companies for free, as long as they provide 'training data' to refine them.

Photo by Mariia Shalabaieva on Unsplash