DigiNews

Tech Watch Articles

← Back to articles

Scent, in Silico

Quality: 8/10 Relevance: 9/10

Summary

The piece surveys efforts to digitize smell using AI, tracing a path from SMILES encoding to modern ML models that learn odor representations. It covers major research programs such as DREAM Olfaction and Google Brain/DeepMind work, and industry efforts like Osmo and Carto, discussing potential applications in fragrance design, safety, and sustainability, as well as the challenges of modeling complex odor mixtures and regulatory considerations.

🚀 Service construit par Johan Denoyer