.png)
When asking questions about how AI could change medicine through biotech, the highest hopes imagine a future where computers generate miracle medicines with the click of a button. While the picture this paints may be enticing, the current reality is a bit more complex. The task of bringing innovative medicines to patients faster is requiring engineers and researchers to rethink more than just what happens in the phase of early drug discovery, and address multiple processes across a complicated and highly regulated industry, from clinical trial recruitment, regulatory submissions, manufacturing, and more.
David Reese believes that AI’s greatest impact will come from improving the thousands of human-led decisions. If Reese, the chief technology officer at Amgen, a leading independent biotech company focusing on fighting some of the world’s toughest diseases, is right, AI could impact every step during the currently 10- to 14-year-long path from discovering a potential new medicine to delivering a medicine to a patient.
“Once we’ve identified a target disease and designed a molecule to address it, we’re only at the beginning of the complex journey to a medicine, and roughly 90 percent of them will fail along the way, according to the National Institutes of Health,” he says. “Our business is fundamentally about making decisions under uncertainty. If we improve decision-making even marginally, the impact could be enormous.”
That’s because drug development moves slowly. Technology evolves quickly. Reese describes this tension as “operating at the intersection of two clocks: one that ticks by in decades, and one that runs on days, weeks, and months.” Living in this intersection are patients living with very real, serious illnesses and unmet medical needs.
Prior to becoming Amgen’s first CTO, Reese held another top position at the company as head of research and development. His background offers a unique opportunity to shape the future of biotechnology through a deep understanding of R&D, coupled with a passion for utilizing AI across the company.
In his prior role, he foresaw the arrival of a “hinge moment,” when computing power and vast amounts of data would become powerful enough “to fundamentally reshape biotechnology and permeate every corner of our business,” Reese says. “Not just R&D, but discovery biology, target biology, manufacturing, and all facets of commercialization.”
He saw that Amgen needed to navigate this convergence of technology and biotechnology proactively, starting by centralizing its technology teams and hiring talent outside the usual biotech recruiting pool.
That started with his first major hire. Sean Bruich, who now leads AI and data at Amgen, spent his career turning big data into measurable impact for household-name corporations. “In consumer technology, better data means better efficiency,” Bruich says. “In biotech, that can mean delivering life-changing medicines to patients faster.” Among the AI and data projects now led by Bruich is one focused on recruiting experienced technologists from digital-native companies who want to apply their expertise to solving some of the toughest problems in human health.
The opportunity ahead, Bruich says, is to master “the art and science of applying new artificial intelligence and machine learning algorithms to Amgen’s collection of biological and operational data in a way that helps us deliver on our mission to serve patients.”
At the heart of Reese’s strategy is the focus on “a handful of initiatives that could move the needle for our patients and the company,” he says.
One priority is the manufacturing of biologic medicines. In biomanufacturing, complex biologic medicines are produced under tightly controlled conditions, where the FDA requires switching equipment configurations between production runs, which can slow output. AI tools are on track to help Amgen reduce the duration of certain changeovers by as much as 36 percent.
This example of an improvement in efficiency helps ensure a reliable supply of medicines for patients who depend on them for ongoing treatment.
Another area of focus is clinical trials, a critical step in the commercialization process. Patient enrollment—identifying the right patients who may qualify for a trial and reaching them in a timely manner—is a common bottleneck to bringing medicines to market. By using AI tools, Amgen has been able to enroll participants up to three times more efficiently in certain trials, helping to shorten the overall timelines for the development of new medicines.
AI-generated algorithms are also helping Amgen predict the optimal viscosity of new medicines, a vital characteristic that affects whether a biologic medicine can be injected easily and manufactured at scale. Determining optimal viscosity was previously a more manual process that can now be completed faster.
None of these changes are "big bang moments,” Reese says. “But they represent improvements across a complex system that ultimately helps improve the speed and reliability of bringing innovative medicines to patients.”
The vision emerging from this strategy is incremental improvements that add up over time.
Drug development in biotech has always existed with uncertainty, and still will, even with AI. Companies like Amgen will continue to design, develop, and test potential medicines that won’t pan out. But improving how decisions are made, from which molecules to pursue, to identifying who qualifies for enrollment in clinical trials, can help shift the odds.
“For patients, this isn’t about algorithms or efficiency,” Reese says. “It’s about time. If we can help them get the right medicine sooner, with fewer setbacks along the way, then all of this work with AI will have mattered.”
