The dataset has nearly doubled in size since 2023, with recent growth driven partly by expanded coverage in microbial and fungal materials alongside advances in profiling and data analysis, according to the company.
The announcement comes amid broader industry interest in AI-assisted ingredient discovery, with companies like Nuritas positioning AI platforms as tools to accelerate the identification of bioactives and early-stage product development.
Expanded material coverage broadens discovery opportunities
Brightseed said the larger dataset allows its platform to screen more candidate compounds while increasing the range of source materials available for analysis.
Speaking to NutraIngredients, Lee Chae, PhD, Co-founder and CEO of Brightseed, said “his growth was fueled by expanding our material coverage into the microbial and fungal spaces in addition to new advances in our profiling and data analysis techniques providing both additional breadth and depth, [and] it allows Brightseed to discover more candidate bioactives, from more source materials, with higher precision and speed than before.”
According to Chae, the expanded dataset also gives partners greater flexibility to evaluate ingredient opportunities in the early stages of product development.
“For partners this means better ability to identify differentiated ingredients, evaluate more opportunities in the innovation funnel, and focus developing most commercially meaningful leads,” he explained.
Brightseed said its platform currently supports modeling across more than 23 health areas, including applications in nutraceuticals and consumer health. The company also said one consumer health partner is using the platform to model pathways associated with GLP-1 signaling.
Company focuses on development efficiency
While AI companies often highlight dataset size, Brightseed said its focus is on improving efficiency during ingredient discovery and development.
“For us, productivity means improving efficiency and confidence of high-potential discoveries moving from broad exploration to actionable decisions,” [which] includes mapping out tractable solution space, screening more candidates, and directing resources toward candidates most likely to succeed scientifically and commercially," said Chae.
The company reported that discovery costs have declined by more than 98% since 2018, which it attributed to investments in automation, proprietary data generation, and AI infrastructure.
“Technically, those gains have been driven by years of investment in proprietary data generation, automation, and AI applications at scale,” Chae said.
“For manufacturers and suppliers, this translates into a more scalable, lower-cost way to identify ingredients with strong scientific and commercial potential,” he added.
Platform positioned as alternative to public datasets
Brightseed said its dataset differs from public compound databases because it links natural compounds to biological mechanisms and human health outcomes tied to product development.
“Most public databases are fragmented, incomplete, or not designed to connect natural compounds to their biological relevance in a way to support product development,” Chae noted.
“General-purpose AI can help interpret information, but they are only useful as the underlying data,” he added.
The company said the platform is designed to help innovation teams assess scientific relevance, source-material potential and differentiation opportunities earlier in the R&D process.
“The platform helps innovation teams generate stronger early signals around scientific evidence, source-material potential, white space, and differentiation, so they can make go/no-go decisions sooner,” Chae said.
Brightseed confirmed that future dataset expansion will focus more heavily on customer product pipelines and commercial priorities.
“The future growth of our dataset will be more targeted to customer value as we are focusing on materials specific to their product goals and commercial pipeline,” Chae said.
