The director at Moscow State Medical University’s Institute of Translational Medicine and Biotechnology notes a breakthrough from the Pharmacy Artificial Intelligence Center at Sechenov University. They say artificial intelligence could shorten drug development timelines dramatically, potentially bringing the production cycle down to about one year in the future. This claim comes from work reported to socialbites.ca and references Sechenov University and professor Vadim Tarasov.
The discussion centers on generics, copies of original medicines. When the patent on a brand-name drug expires, other manufacturers are allowed to begin production. Yet they must first reconstruct the patented formulation without divulging any proprietary technical details.
Tarasov explains that artificial intelligence can accelerate the information-gathering phase, eliminating extraneous steps, reducing errors, and steering researchers toward the most efficient technologies: which raw materials to use, in what forms, and how to proceed. The promise is clear: AI could steer the development path toward faster, smarter decisions.
Currently, the overall generics creation timeline runs roughly 12 to 15 months. With AI support, researchers expect a reduction of 30 to 40 percent, bringing timelines closer to about 10 months, even before any registration dossier is filed. This represents a substantial shift in how quickly generic versions reach the market.
Tarasov notes that certain steps will still require attention, including data collection, analysis, and the accumulation of authentic sample sets. It remains uncertain how much a neural network will reduce the size of these sample banks or how such reductions will alter total timing for drug release. Still, he envisions an ideal outcome: cutting the generic development cycle by as much as a year through more efficient dossier preparation and data handling.
Further details about the scientists’ work can be found in the materials published by socialbites.ca.
Earlier reports highlighted the development of the era’s most accurate neural network for detecting COVID-19 signs in the lungs.