A panel of AI experts explored how artificial intelligence is transforming pet food operations at the U.S. Feed Industry Association’s Pet Food Conference, “Step in to the Future: An AI Journey from Recipe Design to Kibble Production,” held during the 2026 International Production and Processing Expo (IPPE) in Atlanta, Georgia, USA.
Host Dr. Eric Altom, Director of Companion Animal Innovation and Technical Nutritionist at Balchem Animal Nutrition and Health, moderated the discussion with Hana Bielauskas, Senior Vice President and Partner at Inspire PR Group. Johanna Ballesteros, animal science and AI thought leader at SWARM Engineering. Filip Snauwaert, Solution Architect at BESTMIX Software. and Tara Zedajko, Chief Scientific Officer of Ollie Pets.
Here are 10 takeaways from the AI panel discussion.
1. Don’t let data preparation delay your AI adoption
One common misconception is that businesses aren’t ready for AI because their data isn’t perfect. “In my experience approaching customers who are working on use cases, one of the common threads is, ‘The data is not ready. The data is not ready for AI,'” Ballesteros said. “But typically if you delay it or think the data isn’t ready, you’re also delaying the value.”
2. AI will enhance expertise, not replace jobs
Concerns that AI will eliminate the role of humans are misplaced, the panel said. Today’s AI applications are designed to augment missing expertise and increase the efficiency of today’s workforce. Zedeiko explained: Olly, AI can help expand very limited expertise. “If you think about veterinary nutritionists, for example, there are fewer than 100 of them in the country. If there were more, AI probably wouldn’t be as useful.”
3. Extrusion process optimization delivers measurable results
The real-world implementation of AI in pet food extrusion has seen significant operational improvements. Snauwaert shared the results of a proof-of-concept project. “We quickly found that when a new product was produced, the extrusion process started. We were able to start up very quickly.” The company achieved up to a 33% reduction in rework and a 50% reduction in moisture fluctuations.
4. Trust and accuracy are key barriers
For pet food manufacturers, AI predictions must be completely reliable, as errors can impact product safety and quality. “Pet food producers can’t afford one mistake because suddenly the pet food they’re making is no longer viable or healthy for the animals,” Snauwert explained. Data security and maintaining proprietary recipe information also create significant trust barriers.
5. Start with a clear and focused use case
Successful AI implementation starts with identifying specific problems where AI can deliver a measurable return on investment. Experts recommend focusing on processes that require a high degree of human expertise or can significantly improve efficiency, such as production planning, quality assurance, and formulation optimization.
6. System integration solves the problem of data silos
Many pet food companies struggle with data locked in separate systems. “It is very important that these systems and solutions integrate with existing data,” Ballesteros said. By connecting formulation software, production systems, and quality control platforms, AI can optimize across operations rather than individual steps.
7. Ensure safety and compliance with domain-specific AI
For regulatory and safety-critical applications, AI systems should be limited to verified domain-specific information. The American Association of Feed Control Officials (AAFCO) uses an AI assistant called Ava. AAFCO Source material, maintain proper attribution, and prevent misinformation while ensuring accurate regulatory guidance.
8. Simple applications provide quick entry points
Businesses can start experimenting with AI through low-risk applications such as summarizing reports, generating content ideas, analyzing consumer sentiment from reviews, and identifying patterns in product feedback. These applications help teams become familiar with AI capabilities before tackling more complex manufacturing and formulation challenges.
9. Marketing applications require ongoing training
The pet food industry is developing expertise in AI-powered marketing tools. “Many people still don’t have effective training,” Bieliaskas says. “That is changing as AI has become a huge priority in communications.”
10. Photo-based health screening enables personalized nutritional management
AI-powered image analysis makes personalized pet nutrition more accessible. Zedeiko explained: ollie We use photo submissions from pet parents. “Members can take photos of their dogs, including their stool, skin and coat, body condition, teeth, etc. Then, through AI and expert assistance, we provide a personalized health plan.” The company has processed more than 100,000 photos of more than 54,000 dogs to date.
