Food System Innovations Launches the Food Intelligence Lab to Advance AI for Sustainable Protein Development

Food System Innovations (FSI) today announced the launch of its new Food Intelligence Lab to develop the open-source infrastructure needed to accelerate AI-driven sustainable protein development. The Food Intelligence Lab’s goal is to shorten commercialization timelines for better-tasting, more sustainable products.

Sustainable proteins are essential to decarbonizing food systems, which generate about 26% of global greenhouse gas emissions, and livestock supply chains make up more than half of that, contributing 14.5% of all global anthropogenic emissions. Livestock are also responsible for 44% of global methane emissions and 53% of nitrous oxide emissions, two especially potent warming gases.

Reducing animal protein consumption is critical to decarbonizing food systems, but U.S. consumers have been slow to adopt sustainable alternatives into their daily diets. Taste remains one of the sector’s biggest barriers. In NECTAR’s Taste of the Industry 2026 report, just 33% of participants rated dairy-free products as foods they “like” or “like very much,” roughly half the rate of the animal benchmarks, with similarly low overall liking rates for meat alternatives.

More R&D is needed to improve the sensory performance of these products, but many companies face limited budgets, fragmented datasets, and long development cycles that constrain meaningful progress. AI has the potential to dramatically accelerate product development, but the sector has lacked the integrated data infrastructure needed to reliably predict key consumer outcomes like taste and texture. Until now.

Introducing: FSI’s Food Intelligence Lab
The Food Intelligence Lab will generate and curate large-scale datasets, including sensory data from NECTAR and instrumental measurements of sustainable protein products, such as TPA (texture profile analysis), pH, and shear tests, to establish public benchmarks for sensory prediction and formulation design. It will also develop and open-source models that improve product design and prediction tasks, while working with companies, nonprofits, and academic labs to translate those tools into practical applications.

“AI is already transforming fields like drug and materials discovery, but food still lacks the shared infrastructure needed to fully unlock the potential of AI in this space,” said Anna Thomas, Director of Machine Learning at the Food Intelligence Lab and a computer scientist at Stanford University. “We’re building tools to help food scientists iterate faster and create truly exceptional sustainable protein products.”

The Food Intelligence Lab is demonstrating early success through a collaboration with Proxy Foods ΑΙ, a leading AI-powered formulation company. Leveraging Proxy Foods AI food scientist agent, the teams co-developed an optimization system, EGBO (Expert-Guided Bayesian Optimization), that improved the sensory performance of a plant-based Greek-style yogurt by 29% in just 10 formulation iterations over five days. The AI-optimized yogurt matched the animal-based benchmark on three of four key sensory attributes, namely consistency, creaminess, and tanginess. In a comparison with a professional human food scientist given the same time budget, EGBO achieved near-perfect utility of 0.992, vs. 0.850 for the food scientist, ultimately reaching a stronger formulation, faster.

“Food scientists shouldn’t have to spend months on trial-and-error to get texture, mouthfeel, flavor, and aftertaste right. At Proxy Foods AI, we’re building the technologies and operating infrastructure that help companies hit their R&D and sustainability targets faster, with the goal of eventually reaching 90% in silico and reserving the wet lab for the final 10% of validation and optimization,” expressed Panos Kostopoulos, Founder and CEO of Proxy Foods AI. “Every iteration, successful or not, adds value and compounds. Partnering with FSI’s Food Intelligence Lab to open-source these tools is how we accelerate those breakthroughs and ultimately change how we feed the planet for the better.”

Last year, FSI received $2 million from the Bezos Earth Fund to advance sustainable protein product development with AI. By combining human supervision, scientific literature, experiment history, simulators, and foundation models, FSI’s Food Intelligence Lab is building an engine that recommends the next best experiment to try when developing sustainable protein products. Over time, the Food Intelligence Lab aims to help establish a more open, collaborative foundation for sustainable food innovation — enabling researchers, startups, and companies to build products that better meet consumer expectations on taste, texture, and affordability.

In addition to building the EGBO system for formulation design, the Food Intelligence Lab is also tackling sensory prediction, given the time and cost associated with running sensory panels. Food Intelligence Lab researchers recently developed the TasteBench benchmark and competition, with both food-level and molecular-level prediction tasks available publicly on Kaggle. They evaluated existing foundation models and other baseline methods for predicting sensory similarity to the target animal product, and found that the best AI model is competitive with the median individual NECTAR panelist (66.1% pairwise ranking accuracy for the model vs. 65.0% pairwise ranking accuracy for NECTAR panelists).

“We believe AI can be a powerful accelerator for climate and nature solutions when it is paired with the right data, collaboration, and real-world applications, moving promising ideas into impact,” said Dr. Amen Ra Mashariki, Director of AI at the Bezos Earth Fund. “We’re proud to support Food System Innovations’ work, a great example of the kind of opportunity the AI Grand Challenge sought to discover: bringing AI, open science, and sustainable protein research together to help unlock faster progress at scale.”

To learn more about the impact AI can make on sustainable food systems, read the Food Intelligence Lab’s publications and preprints here. The EGBO and TasteBench papers will both be presented at the AI for Scientific Discovery Workshop at the 2026 International Conference on Machine Learning.

To see how FSI’s research can support your organization’s product development, or to participate in pilots or collaborations, connect with the team here. Enter the TasteBench competition by submitting models here for the food-level task and here for the molecular-level task.

About Food System Innovations
Food System Innovations (FSI) is a philanthropic impact platform advancing science, markets, and society to improve food systems. FSI is a strategic hub that accelerates change by combining multiple tools and approaches, acting as both a funder and operator, directly implementing programs while empowering others through grants, investments, and capacity building. Learn more at www.fsi.org.

About the Food Intelligence Lab
The Food Intelligence Lab at Food System Innovations is an interdisciplinary initiative advancing AI-powered sustainable protein development. The Food Intelligence Lab builds open-source datasets, benchmarks, and models to improve how the industry predicts sensory performance, formulates products, and evaluates what works at scale. By combining scientific research with real-world deployment, the Food Intelligence Lab aims to accelerate innovation across the sustainable protein ecosystem. Learn more at www.fsi.org/food-intel-lab.

About the Bezos Earth Fund
The Bezos Earth Fund, guided by the belief that Earth is the best planet in this solar system, works to protect and restore the natural world. Working with partners around the globe, we are developing innovative solutions to ensure the planet remains a place we can – and want to – live on. Based in the United States, the organization is led by President and CEO Tom Taylor, under the direction of Chairman Jeff Bezos and Vice Chair Lauren Sánchez Bezos. To learn more, visit: bezosearthfund.org.

Contact Information

Rachel Stoczko

PR for Food System Innovations

Food System Innovations

Contact via E-mail