BioAge Labs is one of the clearest examples in this directory of AI functioning as core infrastructure rather than a marketing label. The company's approach starts from decades-long human cohort studies, tracking real people's biomarkers, health outcomes, and aging trajectories over time, and applies machine learning to find molecular signals that predict who ages faster or slower, and why. That's a meaningfully different starting point than the more common approach of finding an interesting mechanism in mice and hoping it translates to humans.
Because the target-identification process is grounded in human longitudinal data from the outset, BioAge's clinical-stage programs carry a different kind of evidentiary weight than earlier-stage reprogramming or senolytic companies: the targets were selected because they already correlate with real human aging outcomes, not just because they worked in an animal model. That doesn't guarantee clinical success, drug development remains difficult regardless of how a target was found, but it's a distinct evidence profile worth understanding when comparing companies in this space.
This is also a useful case study for the broader question of what "AI drug discovery" actually means in practice: not a single technology, but a family of approaches applied at different stages of the pipeline, from target identification (BioAge's focus) through molecule design and trial optimization. See the AI drug discovery article for the fuller picture.