The ability to "smell" an illness, like an ear infection or Parkinson's, is not about detecting a universal "sick" odor. It is about recognizing a change from an individual's unique baseline body scent. This skill, once used by doctors, highlights the importance of familiarity in using scent for diagnostic purposes.

A key advantage humans will retain over AI is the ability to translate rich, multi-sensory physical experiences—like touch, smell, and memory—into abstract thought and creative insight. This 'last mile of human experience' is not yet transferable to technology.

Unlike LLMs that scrape the public internet, Osmo had to build its scent dataset from scratch. The fragrance industry's secrecy means no public data exists, forcing Osmo to create a massive proprietary collection of 5 million "sniffs," which now serves as its primary competitive advantage.

Concepts like good taste or judgment aren't magical human traits but are a form of "embedded measurement" in our brains. This data, collected through unique, lived experiences (especially edge cases), is not yet digitized and thus remains a key differentiator from AI models trained on public data.

Vision, a product of 540 million years of evolution, is a highly complex process. However, because it's an innate, effortless ability for humans, we undervalue its difficulty compared to language, which requires conscious effort to learn. This bias impacts how we approach building AI systems.

Early AI models advanced by scraping web text and code. The next revolution, especially in "AI for science," requires overcoming a major hurdle: consolidating and formatting the world's vast but fragmented scientific data across disciplines like chemistry and materials science for model training.

Current multimodal models shoehorn visual data into a 1D text-based sequence. True spatial intelligence is different. It requires a native 3D/4D representation to understand a world governed by physics, not just human-generated language. This is a foundational architectural shift, not an extension of LLMs.

The efficacy of cancer-detecting dogs lies not in identifying a single biomarker but in recognizing a complex, irregular pattern among thousands of emitted chemicals. This suggests that creating an artificial 'nose' for diagnostics requires modeling complex systems, not just searching for a specific molecule, a task well-suited for AI.

Human intelligence is multifaceted. While LLMs excel at linguistic intelligence, they lack spatial intelligence—the ability to understand, reason, and interact within a 3D world. This capability, crucial for tasks from robotics to scientific discovery, is the focus for the next wave of AI models.