Osmo’s AI Skin Irritation Test Points to a Future Without Animal Testing

Landmark AI study The study, published in Alternatives to Laboratory Animals, is described as the first time an AI model for skin irritation has been trained on data generated entirely from validated human-relevant lab methods rather than animal data.

The collaboration evaluated the skin irritation potential of more than 3,000 chemicals using an in vitro Skin Irritation Test, generating safety data that would otherwise have required up to 19,134 rabbits under traditional approaches.

Osmo founder and CEO Alex Wiltschko said, “With the support of AI, we’ve shown that it’s not only possible to eliminate animal testing for skin safety assessment, but that we can do it faster, more accurately, and with better relevance to human health.” The company plans to use this AI-driven method for all future skin-safety assessments to prioritize both animal welfare and consumer protection.

How the model works The Skin Irritation Test in this project was based on reconstructed human epidermis models, a validated in vitro method already accepted by regulators as an alternative to animal testing. This platform was previously used in a Gates Foundation–funded project to discover compounds that can…