This startup keeps human skin alive for a month to train AI

Four years of quiet work has brought Outer Biosciences into the open. The Massachusetts-based startup has been sustaining surgically discarded human skin in a living state for up to 30 days, using the biological data gathered from those samples to train an AI model that can predict the efficacy of new compounds. The company has secured roughly $23 million in funding and now operates with a team of 19.

The process starts with skin that would otherwise be thrown away after surgical procedures. Instead of discarding it, Outer Biosciences keeps the tissue viable for an extended period, creating a testing platform that operates far closer to real human biology than traditional lab methods. That living tissue becomes the foundation for a machine learning system designed to identify which molecules are likely to have therapeutic value, cutting down the time and cost associated with early-stage drug discovery.

By maintaining the skin for a full month, researchers gain a longer window to observe how the tissue reacts to different substances. That extended observation period is a key differentiator. Most models rely on static data or short-lived cell cultures, but Outer Biosciences claims its approach offers a more dynamic and realistic picture of how human skin responds over time.

The company is emerging from stealth at a moment when interest in AI-driven biology is surging. Investors have poured money into startups that merge wet lab work with computational power, and Outer Biosciences fits squarely into that trend. The $23 million raised so far will support further development of its platform and expansion of its team.

With 19 employees, the startup remains lean, but its ambitions are not small. The goal is to build a reliable, scalable way to test compounds without relying on animal models or oversimplified synthetic alternatives. If the technology delivers on its promise, it could reshape how researchers approach dermatological treatments and beyond.