Osmo AI, Founded by Google Brain's Alex Wiltschko, Achieves Human-Level Olfactory Intelligence for Scent Creation
Osmo, a New York startup founded in 2023 by Alex Wiltschko, claims its olfactory foundation model has achieved "human-level olfactory intelligence," recognizing and distinguishing odors as well as an average human and creating new scents at the level of a junior perfumer. This technology is already being used in the fragrance industry, where it has outperformed in-house perfumers of one of the world's largest consumer goods companies. For broader context, explore our Top 100 AI Tools.
The Science Behind AI's Sense of Smell
Osmo's groundbreaking foundation model was developed using an extensive dataset, comprising over 10 million human odor descriptions and 800 million "Scent Tokens." This vast amount of data was meticulously collected by approximately 70 trained data specialists and continuously operating chemical sensors. The process involves a sophisticated "Smell-Printing" robot system, comparable in size to a school bus, which precisely mixes fragrances according to specific formulas. Users can even initiate scent creation through ChatGPT-style prompts, making the process highly accessible and intuitive.
From Digital Data to Olfactory Intelligence
The core of Osmo's innovation lies in its ability to digitize odors and represent them using advanced algorithms, including graph neural networks. This approach allows the AI to understand the complex relationships between molecular structures and their perceived scents. By learning from millions of human descriptions, the model has developed a nuanced understanding of how different chemical compounds translate into specific smells, enabling it to both identify existing odors and predict novel aromatic combinations.
Transforming the Fragrance Industry and Beyond
The immediate application of Osmo's AI is in the fragrance sector, where it is already proving its worth. The ability to rapidly generate and refine new scent profiles offers a substantial advantage, potentially accelerating product development cycles and introducing entirely new categories of fragrances. This capability to create new scents at the level of a junior perfumer suggests a future where AI acts as a powerful co-creator alongside human experts.
Future Implications and Broader Applications
The potential applications of Osmo's olfactory AI extend far beyond perfumery. The company envisions future uses in critical areas such as explosive and hazard detection at airports, where the AI could identify dangerous substances through their unique scent signatures. Furthermore, the technology holds promise for medical diagnosis, by analyzing breath and body odors to detect early signs of diseases. This expansion into diverse fields highlights the versatility and significant societal impact of advanced olfactory AI.
Why This Matters Now
The development of AI with human-level olfactory intelligence marks a significant milestone in artificial intelligence. It demonstrates that AI can master complex sensory tasks traditionally considered exclusive to biological systems. This breakthrough not only offers practical benefits for industries like fragrance and security but also opens new avenues for understanding and interacting with the chemical world. As AI continues to evolve, its ability to perceive and interpret subtle environmental cues will become increasingly vital for a wide range of applications, from enhancing consumer products to safeguarding public health.
Conclusion
Osmo's achievement in developing an AI model capable of human-level olfactory intelligence represents a compelling advancement in the field of artificial intelligence. With its immediate impact on the fragrance industry and promising future applications in security and healthcare, this technology underscores the growing potential of AI to augment human capabilities and solve complex real-world problems. As the model continues to refine its "sense of smell," we can anticipate further innovations that use this unique form of AI perception.
Sources
- Multimodal Data Storage and Retrieval for Embodied AI: A Survey
- Foundation Models for Discovery and Exploration in Chemical Space
- This smell startup says its AI nose is as good as a human’s | MIT Technology Review
- How AI infiltrated perfume | The Verge
- This Startup Is Using AI to Unearth New Smells | WIRED
Recommended AI tools
Google Gemini
Conversational AI
Your everyday Google AI assistant for creativity, research, and productivity
ChatGPT
Conversational AI
AI research, productivity, and conversation—smarter thinking, deeper insights.
Perplexity
Search & Discovery
Clear answers from reliable sources, powered by AI.
Claude
Conversational AI
Your trusted AI collaborator for coding, research, productivity, and enterprise challenges
OpenClaw AI Agent
Productivity & Collaboration
The AI that actually does things.
Cursor
Code Assistance
Your coding agent for building ambitious software
About the Author

Albert Schaper is a co-founder of Best-AI.org. He focuses on product strategy, AI adoption, practical tool selection, and educational content that helps users compare AI products with clearer context.
More from AlbertWas this article helpful?
Found outdated info or have suggestions? Send us a note.