Novi, the platform that helps brands and retailers ensure AI assistants like ChatGPT and Gemini can find and recommend their products in response to shoppers’ unique queries, announced it has filed a provisional patent application with the United States Patent and Trademark Office. The filing covers a proprietary methodology that extracts and maps how large language models select and rank products within a category.
The patent-pending system underlies Novi’s AI Intelligence product. Its foundational insight is that large language models organize meaning around semantic neighborhoods, clusters of related concepts that capture an enormous range of consumer phrasings. Novi’s methodology extracts and maps those neighborhoods, revealing how an LLM actually researches and selects products within a category. That intelligence gives brands competitive positioning they can act on with precision, regardless of how a consumer phrases their question.
“This methodology gives brands a real-time, reliable map of how agents and AI shopping assistants are evaluating and selecting products in their category, turning AI discoverability from guesswork into a discipline,” said Kimberly Shenk, CEO of Novi. “The unit of meaning inside an LLM is a semantic neighborhood and we built this methodology to model those neighborhoods in a way brands can measure and translate into action. When you understand how a model thinks about a product category, you can influence when your products appear regardless of how the prompt was worded.”
Novi’s AI Intelligence product gives consumer packaged goods and retail brands visibility into the category landscape an AI recommendation engine constructs for a given query, the factors associated with which products are selected, and where competitive positioning can be deliberately navigated rather than guessed at. The provisional application covers the core extraction and mapping methodology that powers those views.
As AI-enabled shopping assistants and agents become a primary discovery surface for consumers, Novi’s patent-pending approach gives brands intelligence they can use to evaluate and act on that understanding with greater precision.
About Novi
Novi is a technology platform that helps consumer brands drive discoverability, sales and trust by optimizing their product data so AI shopping tools can find and present it to shoppers. Serving as the bridge between brands, certification organizations and major retailers like Target, Ulta and Sephora, Novi ensures SKU-level data is accurate, consistent and structured, so AI tools can easily surface it, no matter how consumers phrase their shopping questions. Novi then circulates each brand’s optimized digital data footprint, increasing discoverability through consistent citations across platforms. Founded in 2020 and headquartered in Mill Valley, CA, Novi is backed by Tiger Global, Defy, Greylock and other leading venture firms.
To learn more, visit NoviConnect.com.
View source version on businesswire.com: https://www.businesswire.com/news/home/20260623412341/en/
Media gallery
