Scientists at the Wageningen University & Research (WUR) in the Netherlands have received a US$50,000 grant to develop an open-source AI model for improving the taste of plant-based dairy and meat substitutes. The model, which the team says has never before been developed, maps the elements of texture in food building blocks and goes beyond existing AI tools that predict recipes.

The tool aims to understand how molecular components influence texture to enhance consumer appeal as they continue to look for healthier and more sustainable food options.

“Consumers indicate that they often find meat and cheese substitutes too dry or too rubbery,” says Dr. Jack Yang, assistant professor of Physics and Physical Chemistry of Foods at WUR, who is working on the project.

He believes AI models can play a major role in solving such texture complexity-related issues and provide a reference for food producers to create food textures that consumers “really like.”

Tapping tech for enhanced consumer appeal

WUR’s ongoing efforts at improving alt-meat and dairy include an AI-based screening method, Flavour-AI, which it developed last year to address undesirable beany or oxidation flavors in plant-based dairy and meat substitutes.

The tool integrates AI to streamline fermentation, reducing risks, and enhancing the overall consumer experience.

once the latest AI model is developed, Yang and his colleagues plan to make it accessible to food producers and researchers worldwide.

“It really has to become a model for the entire food technology community. In this way, it can be filled with as much relevant data as possible, which means that the development of tasty plant-based alternatives can be accelerated,” Yang emphasizes.

Driving climate-smart innovations

The project, “OLiMPuS,” has received the grant as part of the first phase of the AI for Climate and Nature Grand Challenge hosted by the Bezos Earth Fund, initiated last year to support “bold, AI-powered ideas” that foster sustainability.

“Climate change and biodiversity loss are among the most pressing challenges of our time, and AI has the potential to significantly accelerate our response,” says Dr. Amen Ra Mashariki, director of AI and Data Strategies at the Bezos Earth Fund.

The initiative aims to bolster collaboration between frontline environmental organizations and AI experts for “catalyzing the next wave of transformative solutions,” he adds.

In the second phase, the organization will selec up to 15 projects for implementation grants of up to US$2 million each to scale their solutions over two years.

“The selected projects will use AI to test the possibility of what was once unimaginable, from turning food waste into reusable proteins for food producers to monitoring wildlife poaching through audio recordings,” says Lauren Sánchez, vice chair of the Bezos Earth Fund.

Besides WUR, Cornell University’s College of Agriculture and Life Sciences (US), Delft University of Technology (Netherlands), King’s College London (UK), and the University of Leeds (UK) are some other organizations selected for the first phase grant.