From molecule to market in less time: how AI is transforming pharmaceutical research

An algorithm capable of sifting through thousands of molecules in just a few minutes: a task that would take a human team weeks.

This is not science fiction, but the everyday reality of “Augmented R&D”, the approach in which Artificial Intelligence becomes a computational extension serving researchers and engineers.

According to AIFA data, today 62% of pharmaceutical companies integrate AI solutions into their research and development activities, with a projected growth of 45% over the next five years and a market advancing at an average annual rate of 40%.

The engine of this acceleration is the ability of algorithms to screen vast libraries of compounds, recognize hidden patterns in biological data, and predict the toxicity profiles of drug candidates through computer simulations.

Technologies such as deep learning and generative models allow for the evaluation of millions of molecules in just a few hours, making experiments more targeted and preventing costly false starts.

The paradigm shift can reduce the time it takes for a drug to reach the market by 30%, increase productivity by 40%, and cut engineering costs by 25%.

By limiting “dead ends”, predictive models eliminate molecules with unfavorable profiles right from the start, even before proceeding to chemical synthesis.

Major pharmaceutical multinationals are already investing significant sums to ride the revolution. Isomorphic Labs, a company of the Alphabet/DeepMind group, has signed agreements worth about 3 billion dollars with giants like Eli Lilly and Novartis, recently expanding collaborations with Johnson & Johnson as well.

In parallel, the Recursion platform works closely with Sanofi, Roche, and Bayer; the agreement with Sanofi alone covers up to 15 programs in oncology and immunology, with over 130 million dollars already agreed upon.

Globally, estimates project the segment of AI applied to drug discovery to reach 13.7 billion dollars by 2033.

The phenomenon has not gone unnoticed by the regulatory authorities, engaged in defining new safety frameworks. The FDA has reviewed over 500 practices containing AI components since 2016 and is proposing updated control structures; in Europe, the EMA has published a strategic document to guide developers. Innovation also touches clinical trials, where algorithms optimize patient recruitment and dosage definitions.