
NEW YORK/LONDON — Artificial intelligence is moving deeper into pharmaceutical research as some of the world’s largest drugmakers report that the technology is helping them design molecules, identify promising candidates and assess potential safety problems more quickly. GSK, Bristol Myers Squibb, Novartis, Eli Lilly, Novo Nordisk, AstraZeneca and Sanofi are among the companies expanding AI use beyond administrative tasks and into early-stage drug discovery. Executives told Reuters that the technology has reduced the time needed for some molecule-design and testing cycles, with a few companies reporting substantial improvements in the speed of early research. Sanofi said an AI-enabled approach helped it save at least a year during development of its asthma treatment candidate lunsekimig by supporting earlier clinical decisions and avoiding a dose-selection study. Novartis said AI tools helped researchers identify five targets prioritised for kidney-disease drug development. Reuters’ November 20 report also detailed how Novo Nordisk said AI reduced the number of design, manufacture and test cycles for new molecules from about eight to three, while shortening each cycle from roughly eight weeks to four. The companies’ accounts suggest that AI is becoming a working part of research pipelines rather than simply a technology being tested at the margins.

The progress does not mean AI-designed medicines are already reaching patients faster or proving more effective than drugs developed through conventional research. Many AI-enabled programmes remain in early development, and the transition from a promising molecule to an approved medicine still requires extensive laboratory work, human clinical trials and regulatory review. Researchers cited in the Reuters report cautioned that bottlenecks in trials and approval processes may limit how quickly early discovery gains translate into new treatments. The number of AI-enabled drug candidates entering clinical testing has increased only gradually, despite a growing number of pharmaceutical partnerships and internal research programmes. Companies are also weighing how to work with specialist AI developers while protecting proprietary biological data, which is essential for training models that can make useful predictions. The next stage may connect AI systems more closely with laboratory automation, allowing scientists to design, manufacture and test candidate molecules in a more integrated cycle. HeadlinesXtra has previously covered this emerging field in New Biological AI Model Aims To Accelerate Drug Discovery And Genomic Research. The latest industry update points to wider adoption, but the decisive test will be whether faster early research eventually produces safer, more effective medicines that succeed in clinical trials.