Small molecule therapeutics – generally administered in pill form, as opposed to large molecule drugs, such as biologics – are here to stay. There are times when the industry gives them the cold shoulder in favor of more interesting and innovative therapies, such as CAR T or mRNA. But, with the recent artificial intelligence (AI) craze in the biotech industry, small molecules are back in the limelight – and will be there for at least a large part of this next decade because of the various factors covered below.

The quantity of AI-generated small molecules

Many of the current AI models are very good at generating thousands, if not millions, of small molecule drugs for researchers to parse through. While some mine past literature to iterate on well-known drugs, others analyze protein targets to find chemical matter that can bind them efficiently. Whatever the method, out of the millions of AI-generated drug models, hundreds of small molecule drugs are being added to pipelines at a rapid pace.

A Nature analysis of the impact of AI-contributed small molecules on the biopharma industry showed that the top 20 AI drug discovery biotechs are growing their pipelines by ~36 percent every year. In contrast, the pipelines of the top 20 big pharma companies, which haven’t gone as heavy on AI, remain relatively steady. Another interesting thing to note is the number of programs: In just a few years, the pipelines of these AI biotechs have rapidly grown from 27 small molecule drug candidates in 2017 to more than 158 as of 2021, which is already half as large as big pharma’s slate of 333 drug candidates that year.

The potential quality of AI-enabled drugs

It isn’t just the quantity of small molecules about to hit clinical trials, but the quality as well. One of the major benefits touted by AI biotechs is the ability to tease out chemicals with enhanced properties that lend well to being used as drugs. Some companies, such as Genesis Therapeutics, can even run predictions and simulations of molecules of interest in silico, filtering out obvious duds before moving to animal studies. Even if we assume a large percentage of these drugs won’t make it past the typical clinical development hurdles, there will still be multiple, high-quality small molecule drugs in development that could outshine innovative, but still rough around the edges, therapeutic modalities.

Everyone knows small molecules

These AI-discovered small molecules will be competing with novel therapies (such as cell and gene therapies or mRNA) for similar sets of patients for the same pool of capital. To a risk-averse investor, the tried-and-true small molecule approach is easy to synthesize, and has a built-out manufacturing infrastructure and a well-understood regulatory framework. More importantly, payers are very much aware of the intricacies of this class of medications. This familiarity, alone, gives AI-enabled small molecules an edge over exciting, but unproven, therapeutic approaches.

Investors are excited

We’ve been seeing the same headlines in biotech trade publications since late 2022: AI biotechs are attracting large funding from investors of all kinds. In fact, this niche of the biotech sector is performing very well, while the rest of the industry still lags due to fewer funding sources. With this much financial activity in one sector, and with that sector being solely geared towards small molecule drug development, small molecules are poised to be a mainstay in the industry’s drug development pipeline for the next five years at least.

There will still be a place for new therapeutic approaches in these next few years of small molecule supremacy. There will still be intrepid researchers exploring what we once thought impossible, along with investors willing to take a chance on something that could spark a revolution in medicine. But, for now, small molecules are “in” – at least until generative AI can start generating viable complex proteins and antibodies. When that happens, we'll be back to the days in which we’ll dismiss small molecules as relics of a bygone era.