From shrinking drug development timelines to scouting potential acquisitions, the recent run of first-quarter earnings results marked a shift for Big Pharma from AI hype to explaining how the technology is being applied in practice.
AstraZeneca dedicated a slide in its April 29 earnings presentation to showcasing what it has already achieved—as well as clarifying some bold ambitions. The U.K.-based drugmaker has developed its own open-source generative AI framework, dubbed Reinvent, which has halved the time needed to identify molecular structures that could become potential new medicines.
The Reinvent program is “fundamentally changing the speed at which we can move from concept to lead molecules,” AstraZeneca CEO Pascal Soriot told journalists on an earnings call.
Meanwhile, the company has continued to use its algorithm platform—known as quantitative continuous scoring, or QCS—to identify patients most likely to respond to treatment. QCS is designed to evaluate digitized images of patient tissue samples to quantify targets on the surface of tumor cells as well as inside the cells themselves.
Specifically, QCS is being used to assess non-small cell lung cancer patients who are most likely to benefit from treatment in a phase 3 study of the TROP2-directed antibody drug conjugate Datroway.
When it comes to manufacturing, AstraZeneca hopes its AI Development Agent will ultimately halve chemistry, manufacturing and controls (CMC) development time. The agentic system could shake up synthetic drug development by harnessing simulations, data and in-house expertise.
“AI is transforming our ways of working and is already embedded end-to-end across AstraZeneca, from discovery and development to commercial operations through to healthcare delivery,” Soriot said.
“This is not a future ambition,” the CEO added. “AI is delivering real, measurable impact today, right across our value chain.”
The technology will remain central to AstraZeneca’s push toward its goal of reaching $80 billion in revenue by 2030.
Over at fellow U.K.-based pharma GSK, CEO Luke Miels told journalists that the “number one priority for the usage of AI is the innovation dimension.”
“We’re using it to look at process, cost control and elements like that—but that is typically an off-the-shelf solution,” Miels said on an April 29 earnings call with journalists.
“The proprietary effort, and where we’re putting our best AI people, is on the R&D component, particularly at the earliest stage—the translational part,” he added.
While the company is exploring the use of AI in designing medicines, Miels pointed out that the technology is also being aligned with GSK’s recent M&A strategy of picking up potential rivals to established blockbusters.
“The key thing is looking more broadly at where could that medicine be developed? Where could it be used that maybe the broader world may not have spotted?” he said.
“That becomes very important, not only with our own internal programs, but even more important with business development, because that’s a way that we could identify value that may not be fully reflected in that company or that product,” Miels added.
‘Broadening the use of AI tools’
Over at Bristol Myers Squibb, CEO Chris Boerner, Ph.D., listed “broadening the use of AI tools” alongside greater use of automation in the lab as ways the U.S. pharma is investing in its core R&D infrastructure.
The company has set a target of using AI to halve the time taken to select targets and design molecules while “applying greater rigor so that only the most differentiated molecules advance in late development,” Boerner said on an April 30 earnings call with analysts.
When it comes to the later stages of drug development, BMS is already using AI to streamline clinical operations, shorten development timelines and improve quality oversight over time, the CEO said.
“We expect these efforts to deliver a 30% reduction in cycle times versus just a few years ago,” Boerner added, pointing to a March deal with Faro to scale AI-powered workflows across clinical trial design, drafting, validation and optimization.
BMS wasn’t the only Big Pharma on the hunt for AI-powered partnerships in recent months. Novo Nordisk followed in the footsteps of Sanofi and Eli Lilly by striking its own pact with OpenAI in April, with the ambition of integrating AI “globally from drug discovery to commercial operations.”
Novo’s announcement came days after OpenAI unveiled a new reasoning model, dubbed GPT-Rosalind, specifically designed to support research in biology, drug discovery and translational medicine. The model is now available as a research preview in ChatGPT, Codex and the API for customers through OpenAI’s trusted access program.
Meanwhile, Merck & Co. picked Google Cloud for its own AI partner. As part of an investment valued at up to $1 billion, Merck will get access to the tech giant’s agentic AI platform across its R&D operations, manufacturing, commercial teams and corporate functions.
Speaking to analysts on an April 30 earnings call, Merck CEO Robert Davis framed the Google Cloud deal within the broader context of a partnership with Tempus AI to accelerate the discovery of precision medicine biomarkers, as well as a collaboration to access Mayo Clinic laboratory results, medical imaging, clinical notes and molecular data to support the validation of AI models for drug discovery.
“These efforts support improved productivity across our organization and create a real opportunity to advance the innovation in our pipeline with greater speed and with a higher likelihood of ultimately reaching patients,” Davis explained.
