INNOVATION | SUSTAINABILITY

Food manufacturers have spent decades trying to reduce waste. Artificial intelligence and advanced fermentation are creating a more ambitious possibility: turning factory side streams into valuable food ingredients.

Key Takeaways

• Fermentation is well established. What AI could change is how consistently and economically these processes operate at industrial scale.
• Two major projects are testing the concept. Protein Industries Canada is backing an AI enabled fermentation platform, while CSIRO and the University of Leeds are developing a parallel system for vegetable, grain and dairy side streams.
• Variability is the core technical challenge. Industrial side streams fluctuate in composition, and AI could help processors respond to those changes in real time.
• The economics still have to work. Collection, stabilization, fermentation and purification all cost money, and ingredients must ultimately compete on price and performance.
• The bigger opportunity is value creation, not just waste reduction. A more circular food factory could generate secondary revenue from streams that once left as waste.

Peels, pulp, spent grain, whey and nutrient rich processing streams have traditionally been treated as costs to manage or materials to divert into lower value uses such as animal feed.

That equation may be changing.

Advances in fermentation, sensors and artificial intelligence are creating opportunities to treat some food processing side streams as feedstocks for a second manufacturing process.

Microorganisms such as yeasts, bacteria and fungi can convert components of food processing byproducts into microbial protein, organic acids, enzymes and other functional ingredients. Fermentation itself is well established. What AI could change is how consistently and economically these processes operate.

Making Fermentation More Predictable

One major challenge is variability.

Industrial side streams can change in carbohydrate concentration, moisture, acidity, mineral content and inhibitory compounds. Research published in Current Opinion in Biotechnology highlights this compositional complexity as an important challenge when using food processing byproducts for fermentation.

AI could help manufacturers respond to these variations in real time.

In June 2026, Protein Industries Canada announced a C$1.4 million project involving Crush Dynamics and Atomic47 Labs to develop an AI enabled fermentation platform, with Protein Industries Canada committing C$607,000 and the partners funding the remainder. The technology is intended to combine industrial sensor data with machine learning to assess fermentation conditions, food safety indicators, energy performance and process health.

A separate collaboration involving Australia’s CSIRO and the University of Leeds, backed by the Bezos Earth Fund, is developing an AI platform to optimize fermentation of vegetable waste, grain byproducts and cheese manufacturing residues into microbial protein.

The ambition is significant: make ingredients produced from side streams competitive with conventional alternatives.

From Waste Reduction to Value Creation

This changes the sustainability question.

Instead of asking only how manufacturers can generate less waste, companies can increasingly ask what else they can manufacture from the raw materials they already purchase.

Consider potato processing. Producing fries and other potato products generates peel, starch containing water and other organic streams. If valuable components can be recovered or biologically converted economically, the same potato could generate additional products and potentially additional revenue.

Similar opportunities exist in brewing, dairy, milling, and fruit and vegetable processing.

A 2026 review in Food and Humanity identified food byproducts as potential sources of ingredients, nutraceuticals, biomaterials and bioenergy. The researchers also highlighted AI based monitoring and process analytics as potential tools for improving production consistency.

The Economics Still Have to Work

AI does not automatically make waste valuable.

Collection, separation, stabilization, pretreatment, fermentation, purification and drying all cost money. Food safety, regulatory approval, sensory performance and ingredient functionality create additional hurdles.

This is an important distinction as enthusiasm around AI and circular manufacturing grows.

The scientific breakthrough is not simply demonstrating that food waste can become something useful. Researchers have already shown that biological processes can convert food residues into proteins, organic acids and other valuable compounds.

The harder challenge is making those processes consistent, safe and inexpensive at industrial scale.

AI could help by detecting process deviations earlier, predicting fermentation behaviour and adjusting operating conditions as feedstocks change. But manufacturers will still need to demonstrate that the value of the resulting ingredient exceeds the cost of producing it.

What Food Manufacturers Should Watch

The emerging opportunity goes beyond meeting sustainability targets.

A more circular food factory could manufacture its primary products while recovering or converting side streams into secondary ingredients. Waste management could gradually become intertwined with ingredient innovation, productivity and revenue generation.

For manufacturers, the critical question is therefore not whether AI can find uses for food waste.

It is whether AI assisted bioprocessing can make side stream ingredients commercially competitive with ingredients produced from conventional raw materials.

If that happens, tomorrow’s food factory may be judged not only by how little waste it produces, but by how much value it creates from what once left the factory as waste.

---

Sources

1. Protein Industries Canada. AI enabled fermentation platform for smart, sustainable food manufacturing. June 24, 2026.
2. CSIRO. CSIRO partners on global project to boost food security. January 14, 2026.
3. University of Leeds. Building an AI platform to help food security. January 14, 2026.
4. Current Opinion in Biotechnology. Upcycling food processing byproducts via advanced yeast fermentation. 2026.
5. Food and Humanity. From waste to wealth: Innovating with food byproducts. May 2026.
6. Agriculture. Agri Food Byproducts as Food Tech Material Resources: A Critical Review of Processing, Functionality, Matrix Validation, and Readiness. July 25, 2026.
7. Nature Food. Artificial intelligence for food innovation. 2026.