Closing the allergy information gap: AI portal predicts takeout options based on dietary lifestyle
01 Mar 2021 --- Canada-based start-up Honeycomb.ai is tapping into artificial intelligence (AI) to predict which menu items fit users’ dietary requirements.
The platform could spell an end to the frustrating process of manually sifting through menus by helping users instantly find foods tailored to their needs.
“After my personal diagnosis, I realized how many others struggle with finding food to eat due to a variety of reasons,” says co-founder Tamir Barzilai, who was prompted to create the portal after an ulcerative colitis scare.
“The lack of ubiquitous dietary and ingredient transparency didn’t make sense from both consumer and business perspectives,” he affirms.
Automating the guesswork
Honeycomb.ai predicts dietary information and key ingredients for menu items, expediting the process of finding dishes that are likely to be suitable for consumers’ diets.
The results are shown by city and dietary combination and they constantly improve as the algorithm learns. Input from restaurants, feedback from users and internal training processes collectively help make the predictions more accurate. 
Honeycomb.ai predicts which menu items are likely to be suitable for consumers’ diets.
The platform supports vegan, vegetarian, celiac, gluten-free, ketogenic, paleo, kosher, halal, and low-FODMAP diets, along with the 29 most common allergens and ingredients.
“One of the most challenging aspects of accommodating dietary needs is the different levels of severity,” explains Barzilai.
“Someone might simply dislike mushrooms, or they may be mildly intolerant to dairy, or they may have a severe anaphylactic reaction to shellfish by means of cross-contamination.”
“We’ve built this sense of severity into our software.”
The information gap online
Growth in online food ordering has prompted customers with dietary needs to search online for dietary answers more than ever before, notes the start-up.
With over 52 percent of US consumers following at least one diet, and less than 10 percent of restaurants labeling dietary information – which is typically not exhaustive – the information gap has never been wider, stresses Barzilai.
After testing with over 100,000 unique users in Western Canada, the Honeycomb.ai team is expanding globally.
The first phase is currently available for users in Canada, Australia, and the US, and free to use.
Getting personal
AI’s ability to learn and rapidly sift through information is proving to be a powerful tool in the world of personalized nutrition.
The online platform Qina launched in January, connecting the personalized nutrition industry with a B2B market insight database.
Last May, Nestlé-owned Persona’s CEO flagged that COVID-19 would have a lasting impact on tailored nutrition and growth opportunities in telehealth.
Further supporting the trend, Innova Market Insights’ third top trend for 2021: “Tailored to Fit,” spotlights the rise of personalized nutrition in the F&B industry.
Edited by Missy Green
