A field that is smaller than it looks
The “AI drug discovery companies” category sounds broad, but the field of firms that actually use machine learning to design medicines, as opposed to using it for data entry or trial logistics, is smaller than the marketing suggests. The ones that matter split into a handful of archetypes: the pure-play platform companies, the partnership-driven discovery shops, and the software vendors that sell tools to pharma rather than developing drugs themselves.
The promo behind the “sub-$10 AI drug stock” tease, Alexander Green’s Oxford Club pitch, points at one of the partnership-driven names. The company is Absci, and its model is a good lens for understanding the whole category. Absci converts proteins into a digital form its generative AI can search, then licenses the resulting candidates to pharmaceutical partners. That is the discovery-shop archetype: a small firm with a platform that big pharma pays to use.
The partnership model, with real names attached
Absci’s partner roster is the clearest illustration of how this model works. Merck signed a drug discovery deal in 2022 worth up to about $610 million in milestone payments. AstraZeneca followed in 2023 with a deal worth up to about $247 million to apply the platform to cancer drugs. More recently, Absci struck a hardware collaboration with AMD that came with a $20 million investment in January 2025, and Eli Lilly made a $40 million strategic investment tied to its lead program.
Those are the mechanics in practice: a platform gets validated one partnership at a time, and each milestone payment is a signal that the technology is doing something real. The catch is that milestone payments are lumpy and tied to development progress, so revenue does not rise in a smooth line. We explain the mechanics of that revenue model in our AI biotech stocks piece.
How the field actually breaks down
Beyond the partnership shops, the field includes the software vendors, the fully integrated players, and the big pharma incumbents building their own AI groups. The software vendors sell machine learning tools to drugmakers and collect subscription revenue without taking clinical risk. The fully integrated players discover and develop their own drugs, keeping all the upside and all the risk. The incumbents are the quietest group: most large pharmaceutical companies now run internal AI teams, which means the competitive set for a small platform company includes the very partners it hopes to license to.
That last point is easy to miss. When a small AI firm signs a deal with a pharma giant, it is not just gaining a customer. It is also watching that customer build competing capability in-house. The partnership model works best when the platform has something the partner cannot easily replicate, which is why the quality of the underlying science matters more than the number of press releases.
What separates the real ones from the slogans
The clearest test is whether a company has produced a drug candidate that reached clinical trials. AI is good at generating promising molecules; the hard part is getting one through human testing. Our AI drug discovery explainer covers the mechanism in depth, but the investing version is simple: a company that has a candidate in trials is categorically different from one that only has a platform and a partnership.
Absci passes that test. Its lead asset, ABS-201, is an antibody for pattern hair loss and endometriosis that is moving through trials, with proof-of-concept data expected in the second half of 2026. That clinical asset is what separates it from the dozens of AI drug discovery companies that exist mostly as slide decks. For a framework on evaluating the rest of the field, see our AI drug discovery stocks guide.
The public-versus-private split
One reason the public AI drug discovery universe feels small is that many of the most interesting firms are still private. A large share of the venture funding in this space sits in companies that have not listed, which means public investors have a short list of pure plays to choose from, and Absci is one of the few that combines a generative platform with a clinical-stage asset.
That scarcity cuts both ways. It means the public names command attention when a promotional cycle turns to the space, which is part of what is happening with the “sub-$10 AI drug stock” pitch. It also means there are fewer ways to diversify the theme, so the public pure plays trade more on sentiment than a broad sector would. The lesson for an investor is to treat the public list as a narrow set of high-conviction names rather than a diversified category.
Ready to see the research? Click here to access Alexander Green’s report.
NewsletterVetter is an independent publication. We receive compensation from some of the services we review through affiliate links. Nothing on this site is investment advice. Always do your own research.