With artificial intelligence (AI) continuing to advance within the pharma industry, it is transforming how new molecules are discovered, developed and delivered.
Offering opportunities to accelerate timelines, reduce experimental burden and improve decision-making throughout the drug development lifecycle, Emily Letton speaks to Quotient Science’s Matt Paterson to explore the topic.
From discovery to development
In target identification, for example, AI algorithms analyse extensive datasets to isolate promising drug targets. For molecule design and screening, simulations can predict compound–target interactions, thereby minimising the need for large-scale physical testing.
In 2026, we expect AI to continue to make an impact in different stages of development. Notably, we see the potential of AI to have a substantial role in optimising clinical trials and guiding formulation development.
This will allow development teams to better predict patient responses and improve trial outcomes, as well as select the most appropriate drug delivery methods based on chemical structures and desired therapeutic outcomes.
Reducing trial-and-error in formulation
In formulation development, AI is likely to reduce our reliance on traditional "trial-and-error" methodologies by virtually simulating experiments to identify optimal formulation compositions and manufacturing process parameters before any laboratory or clinical testing.
In partnership with Intrepid Laboratories, a Toronto-based leader in the pharmaceutical formulation science space, Quotient Sciences is exploring this opportunity.
Using Intrepid’s machine learning model ANDROMEDA — the first AI platform designed to develop and optimise the clinical performance of drug products — we are exploring ways to reduce experimental burden and minimise drug substance requirements during product design.
These approaches enable new drug products to be tested more rapidly in the clinic and manufactured more efficiently.
Human expertise will remain essential
While AI advances, the expertise and insight of the person or team applying the tools cannot be understated. It is the role of the scientist or formulator to help provide data for AI/LLM models, frame the right questions, validate outputs and interpret predictions to ensure that AI supports — rather than overrides — human decision making.
It is well-known that the path to bringing a drug to market, on average, takes more than a decade and often several billion dollars.
AI presents an opportunity to accelerate these timelines, reduce the R&D costs and enable more accurate and personalised treatments for patients.
