AI drug discovery grew up at the same conference that gave it an early award a decade ago
Ten years ago, Insilico Medicine received the PMWC Emerging Company Award for applying generative AI and machine learning to drug discovery. At the time, the idea that artificial intelligence could meaningfully accelerate the notoriously slow, expensive process of finding new drugs was more aspiration than evidence.
At the Precision Medicine World Conference 2026, held March 4-6 in Silicon Valley, that same company's founder and CEO, Alex Zhavoronkov, returned as track chair for the Precision Aging and Longevity sessions - and presented the conference's Luminary Award to Greg Brockman, president and co-founder of OpenAI.
From emerging to operational
The trajectory illustrates a broader shift in the field. What was speculative in 2015 is now operational. Insilico Medicine reports that between 2021 and 2024, using its Pharma.AI platform combined with automated laboratories, the company nominated 20 preclinical drug candidates. The average time from project initiation to preclinical candidate nomination was 12 to 18 months, with only 60 to 200 molecules synthesized and tested per program. Traditional early-stage drug discovery typically requires 3 to 6 years for the same milestone.
Those numbers are self-reported and have not undergone independent verification in this context, but they align with a broader trend in the industry where AI-driven approaches are compressing early discovery timelines across multiple organizations.
Measuring biological age with machine learning
Zhavoronkov moderated a session on AI-driven biomarkers for quantifying aging, featuring Steve Horvath of Altos Labs - the researcher who pioneered epigenetic aging clocks - and Mahdi Moqri of Harvard, co-director of the Biomarkers of Aging Consortium.
The discussion centered on how machine learning can be applied to genomic, epigenetic, and clinical datasets to move beyond chronological age and measure biological age - the actual condition of a person's cells and tissues, which can diverge significantly from their birth date. The goal is not just academic. If biological age can be reliably measured, interventions that slow or reverse aging could be tested with a concrete, measurable endpoint rather than waiting decades to see who develops disease.
Epigenetic clocks, which measure patterns of DNA methylation that change predictably with age, have become the most established tools in this space. But challenges remain: different clocks give different readings, biological age may vary across organs within the same person, and the field has not yet agreed on which biomarkers best predict future health outcomes versus merely correlating with past exposure.
AI at the center of biomedicine
The conference more broadly reflected the integration of artificial intelligence into multiple stages of biomedical research. Steve Wozniak, co-founder of Apple, received the Pioneer Award. The fact that the Luminary Award went to OpenAI's president - not a biologist or physician, but a technologist - signals how deeply the field now considers AI infrastructure to be foundational to the future of medicine.
The 2026 meeting also coincided with the 25th anniversary of the first draft of the Human Genome Project, a milestone that underscored how genomics, computational biology, and AI have converged into a single research enterprise.
Caveats worth noting
Conference presentations and awards are not the same as peer-reviewed clinical results. Insilico Medicine's preclinical candidates have not yet produced approved drugs, and the attrition rate from preclinical nomination to market approval historically exceeds 90% across the pharmaceutical industry. Speed in early discovery is valuable, but it does not guarantee that the resulting candidates will prove safe and effective in human trials.
The longevity and aging field, despite genuine scientific progress, remains entangled with hype and commercial interests. Distinguishing between companies that are advancing rigorous science and those leveraging the language of AI and longevity for investor appeal requires sustained scrutiny - the kind that conferences, by their nature, are not designed to provide.