A team of human sniffers at Osmo's lab in New Jersey is creating a vast dataset of odor descriptions, which is being used to train the company's AI model. The goal is to give computers a sense of smell, leveraging the unique power of scent molecules to unlock new possibilities in fields like medicine and beyond. Osmo's model has already shown promise in identifying fragrances, insect repellents, and even counterfeit sneakers. However, critics argue that the company's claims are still largely undocumented and reliant on human judgment. Scientists warn that the AI may struggle to capture the complexity of human smell, which can be influenced by context and subtle differences in molecular composition. Despite these challenges, Osmo's work has the potential to revolutionize the way we interact with the world around us, from medical diagnostics to product development. By cracking the code of human smell, Osmo's AI model could unlock new insights and applications that were previously unimaginable.