Transforming data into knowledge: a GFSI panel examines the role of AI in detecting food safety risks
At the Global Food Safety Initiative conference in Vancouver, industry leaders highlighted how AI and advanced analytics are transforming risk management in food safety. Companies like Ecolab and McCain Foods are applying predictive models to anticipate failures, reduce contamination, and optimize audits in global supply chains. Analytics allows for the identification of emerging risks and the prioritization of interventions in real time. However, it was emphasized that effective adoption requires governance, data infrastructure, and organizational alignment. This trend marks a transition toward predictive systems that strengthen prevention and resilience in complex supply chains.
Food safety officials at GFSI Vancouver say AI and advanced analytics are transforming how companies detect and prevent risks in global supply chains.
Artificial intelligence (AI) and advanced data analytics are rapidly transforming how food companies detect and prevent safety risks, industry leaders told delegates at the Global Food Safety Initiative (GFSI).
As food supply chains become more complex and interconnected, companies are increasingly turning to data-driven tools to analyze operational signals, identify emerging risks early, and intervene before problems escalate.
.
.
Opening a panel discussion on data-driven food safety, Lisa Robinson, Vice President of Global Food Safety and Public Health at Ecolab, stated that the scale and speed of modern food systems are highlighting the limitations of traditional monitoring approaches.
“Food safety is complex—it has never been simple—but it is becoming increasingly interconnected and evolving more rapidly. Risks can manifest earlier, in different places, and sometimes in ways that our traditional systems were not designed to detect.”
Go beyond averages
One area where this shift is already visible is in retail operations, where companies are using AI to analyze large volumes of operational and audit data. Instead of relying on overall system averages, organizations are increasingly turning to AI to pinpoint where risks arise.
Retailers are already applying this approach to analyze the massive amounts of audit data generated across thousands of stores.
Catherine Cosby, senior director of food safety and compliance at The Kroger Co., said the technology is helping teams prioritize where more support is needed.
“We have just over 2,500 stores and undergo regular audits throughout the year. Being able to examine that data, analyze it, and make an informed decision about which stores may need additional food safety support or where we should be doing things differently [is invaluable].”
By identifying critical risk points in advance, companies can intervene more quickly and allocate resources more effectively.
Predictive Safety in the Production Plant
This same shift toward predictive risk management is also transforming the food industry.
The previous reliance on fixed maintenance schedules is becoming increasingly unnecessary, as machine learning can anticipate equipment failures that could lead to contamination incidents.
.
.
Tola Alade-Lambo, Vice President of Food Safety and Quality at McCain Foods, stated that predictive maintenance models are helping manufacturers act before problems arise.
“Historically, our preventative maintenance was based on data like: ‘Last year it failed after six months, or two years ago it failed after seven months.’ Now we create models that tell us when it is about to fail.”
These tools can reduce the likelihood of contamination by foreign materials, while improving operational efficiency across all production lines.
Data: Transforming Conversations Across the Supply Chain
Data is also changing how companies internally address food safety risks in complex global supply chains.
Gary van Breda, Director of Global Food Safety (Food and Packaging Suppliers) and Consumer Product Safety at McDonald’s, stated that visualizing operational data helps align conversations between suppliers, operators, and restaurants.
“When you visualize data to understand the implications and how it interconnects between an operator, a supplier, and a restaurant, for example, it becomes easier to have a conversation about aspects of the business that are much more meaningful.”
By transforming large datasets into clearer information, companies can help teams in different areas of the organization understand how risks propagate through the system.
Technology still requires strategy.
The speakers also cautioned that technology alone will not solve food security problems.
Organizations must also develop the necessary infrastructure, governance, and internal alignment to support digital systems.
Pavlos Fragkopoulos, global director of quality management at Mars Petcare, warned that many organizations are adopting AI tools without clearly defining the problem they are trying to solve and, as a result, are struggling to generate benefits because they lack the necessary infrastructure and internal support to scale them, referencing a recent MIT report.
The report, “The Generative AI Gap: State of AI in Business 2025,” revealed that despite business investments of between $30 billion and $40 billion in generative AI, most organizations are not seeing a quantifiable business return, while only about 5% of embedded AI pilot projects are generating significant value.
The discussion highlighted a broader shift occurring across the sector, as companies move from analyzing historical food safety data to using digital tools that can identify emerging risks earlier and facilitate faster decision-making in complex global supply chains.
Source: inocuidadhoy.com
.
