From the floor of IAFP 2025: AI moves beyond hype to deliver concrete food safety solutions

David Monk, Senior Director of Restaurant Food Safety Culture at Chick-fil-A, addressing attendees at IAFP 2025. Monk detailed how AI tools, such as predictive modelling and machine learning, are actively being used to enhance human decision-making in real-world food safety challenges

At the 2025 International Association for Food Protection (IAFP) Annual Meeting, artificial intelligence was presented as an active tool already delivering benefits across the food sector. During the symposium Cutting through the hype: real-world benefits of AI in food safety, speakers from industry and regulatory bodies shared how machine learning, generative AI, and open AI systems are being used in real-world settings.

The session showcased how AI is supporting regulatory compliance, optimising operations and improving food safety outcomes. Not limited to just future promise, presenters shared examples of AI’s present-day impact.

Predictive modelling drives proactive interventions

Amani Babekir of Ecolab explained how teams are using historical data, including environmental monitoring, internal audits and third-party inspections, to predict compliance levels at retail stores based on health inspection reports. This enables focused intervention to mitigate risk and drive improvement.

She also shared how GenAI is being used for microbial environmental management. When unacceptable swabbing results occur, GenAI generates root cause analyses and corrective actions to improve hygiene programmes.

Cross-sector collaboration through data integration

Brendan Ring of Creme Global discussed AI’s role in supporting data integration across the collaborative but fragmented food supply chain. Creme Global is contributing to a multi-agency FDA platform that consolidates siloed data from federal bodies such as the USDA and EPA into a unified format to support new insights.

He also highlighted the Food Industry Intelligence Network (FIIN), where over 70 member companies anonymously share sensitive product testing data to identify aggregated trends and detect potential issues. Ring described their work with Western Growers, using AI to analyse data from leafy green growers and identify cost-effective preventive measures.

Ring added that AI is capable of extracting and contextualising information from previously inaccessible unstructured data sources, including reports and publications. It is also being used to convert complex legislation into accessible formats such as AI-generated podcasts.

AI supports, not replaces, human expertise

David Monk of Chick-fil-A said AI is being used to help humans make faster, more accurate decisions when working with complex data. Examples include predictive modelling for pathogen growth (such as Salmonella), machine learning to forecast product temperatures during power outages, and Monte Carlo simulations for Norovirus spread and mitigation.

He described Chick-fil-A’s AI-powered decision support tool, Tango, which continuously learns and supports operational decisions. Monk warned about hallucinations in large language models and stressed the need for human oversight to tune models, analyse outputs and set thresholds. Babekir reinforced this, noting that AI “will not eliminate the need for subject matter experts”.

Regulators show active involvement

Mark Moorman from the FDA expressed the agency’s excitement about AI’s potential and highlighted the FDA’s involvement in these technologies, including data sharing agreements with Western Growers.

Sarah Murphy, also from the FDA and an organiser of the symposium, encouraged attendees to engage with IAFP’s Data Management and Analytics Professional Development Group. She noted that she helped to establish the group, and said “anybody here is welcome” to join. She added that the PDG is the place to go “if you want to see more programming like this at IAFP”.