The week in health tech, explained in plain English.Get the free newsletter →
October 23, 2026

Takeaways from the most recent news in the technology and policies shaping healthcare.

Health IT

Bristol Myers Squibb Claims Pharma's Largest NVIDIA AI Supercomputer

By Healthcare Technology News ·

Bristol Myers Squibb has announced it is assembling the largest AI supercomputer in the life sciences industry, built on NVIDIA hardware, according to STAT News. The catch: it is the third drugmaker in nine months to make essentially the same claim, turning "largest" into a marketing badge more than a verifiable spec.

The wave of announcements reflects a broader race among pharmaceutical companies to build in-house AI infrastructure for drug discovery and development. Betting that massive computing power can speed the identification of new molecules and shorten research timelines, drugmakers are pouring money into GPU clusters and deepening ties with NVIDIA, now a central supplier to the sector.

For readers, the signal matters more than the superlative. Competing claims to the same title suggest the boasts are hard to compare, since companies rarely disclose consistent metrics on compute capacity. What is clear is that AI infrastructure has become a competitive differentiator pharma leaders now feel compelled to publicize, whether or not the underlying science has caught up.

Sources

More in Health IT

Health IT

Epic Opens Its Platform Wider to AI and Outside Developers

Epic used its Open@Epic developer conference to open its platform wider to outside AI and apps while reminding developers that the resulting traffic runs on hospitals' infrastructure.

Why it matters: As the dominant EHR vendor, Epic's decision to open its platform shapes how and where clinical AI gets deployed across most U.S. hospitals.

Health IT

After AI Go-Live, Who Owns Monitoring? Often No One

Health systems scrutinize AI before go-live, but post-deployment monitoring responsibilities often remain undefined, leaving no clear owner when models drift.

Why it matters: Unmonitored AI models can quietly degrade in clinical settings, putting patient safety at risk when no one is accountable for watching performance after launch.