Using AI to Develop Better, More Sustainable Food Packaging Faster

By Sunil Sanghavi, CEO of NobleAI

Avocado pits, popcorn, and mushrooms — not items on a shopping list, but rather a few of the items that innovative companies are beginning to use in sustainable food packaging. Industry insiders are understandably, and increasingly, pushing for greener food packaging as awareness of the long-term detrimental impact of certain ingredients in packaging materials continues to grow. One survey of consumer habits revealed that 82% of respondents were willing to pay more for sustainable packaging. In fact, according to reporting on Earth.org , the global sustainable food packaging market is set to grow to $280 billion in value by 2026 , a 50% increase from 2021.?

Traditional packaging materials are designed to enhance specific properties. For example, BPA (Bisphenol A) strengthens materials, and phthalates, used as plasticizers, increase flexibility. PFAs (per/polyfluoroalkyl substances) are other universally popular chemicals that make packaging resistant to water, oil, heat, and degradation — ideal for preserving the product. But as we are increasingly learning, these “forever” materials also make them a threat to human health and the environment, intensifying the urgent search for safe, sustainable, reliable, and consistent replacements. Lawmakers are also sounding warning bells about these ingredients and spurring company action with new regulations related to PFAs and food packaging beginning in 2024.?

As food manufacturers and distributors work overtime to eliminate these chemicals, often experimenting with packaging materials derived from biodegradable materials, they are also facing steep R&D costs. In addition, the traditional, lab-based R&D experimental approach to identifying new materials formulations simply isn’t fast enough to keep up with demand. The science involved in exploring viable replacements of toxic components with eco-friendly alternatives, including bio-based materials is complex, requiring research across a vast space of possible chemical and materials combinations, then evaluating performance and quality in different scenarios and applications, and testing for safety and environmental impact. This time consuming and expensive process involves detailed molecular understanding as well as complex lab-based chemistry experiments. And all of this must be done in a competitive environment where the pressure to get new products to market quickly, affordably, and reliably is intense — no easy feat, especially considering the supply chain challenges that have wreaked havoc in the last few years.

Today, new artificial intelligence (AI) technologies are changing this dynamic, speeding the identification of alternatives without the prohibitive costs and timelines associated with traditional approaches. While some, mostly smaller companies, have invested significant time, money, and creativity into uncovering innovative solutions for sustainable packaging — like Loving Earth, an Australian company that packages their vegan chocolate bars in a compostable film derived from wood pulp and non-GMO corn, or No Evil Foods that sells small batch, plant-based meat alternatives packaged in fully compostable materials printed with plant-based ink — Science Based AI can help companies do this much faster and at scale.?

For data scientists, infusing AI into research provides a means to get to the lab-based experiments with the most promising materials quickly. By running thousands of experiments virtually, saving expensive and time-consuming lab experiments for the most promising candidates, scientists can quickly identify potential compounds and speed the discovery of better performing, safer, and more environmentally sustainable packaging materials. Science-based AI works by training AI models with relevant scientific knowledge, understanding of physical systems, and relevant data sets specific to your company’s and industry’s unique needs. AI for science can predict the behaviors of new materials under a range of conditions, sidestepping the time-intensive empirical approach, and once this is done, scientists can then quickly explore the vast space of possible chemical formulations or combinations, identifying compounds that could replace “forever chemicals” and other toxic ingredients without sacrificing performance or quality.??

By increasing the bio-content of formulations and identifying low-toxicity plasticizers, new science-infused AI methods can play a key role in minimizing leaching from plastics and accelerating the development and market introduction of sustainable materials and additives and therefore, better performing, more environmentally sustainable packaging.

Sunil Sanghavi is CEO of NobleAI , a pioneer in science-based AI solutions for chemical and material informatics. Sunil has a rich operating background in deep-tech companies. Most recently, he was Senior Investment Director at Intel Capital, investing in AI/ML hardware and software companies including Motivo, Untether AI, Syntiant, and Kyndi.