Recommendations for the Development of Artificial Intelligence Applications for the Retail Level

Area of Science:

• Food Safety
• Artificial Intelligence
• Public Health

Background:

• Current artificial intelligence (AI) applications in food safety primarily focus on manufacturing and distribution.
• There is a significant need for AI tools to support food safety practices at the retail level, particularly for outbreak investigations and risk assessments.

Purpose of the Study:

• To highlight the necessity of AI applications for retail food safety, including foodborne illness outbreak investigations, Hazard Analysis and Critical Control Points (HACCP) plan development, and root cause analyses.
• To propose a path forward for developing more robust AI, moving beyond pattern recognition to cognitive models.

Main Methods:

• The study reviews existing AI applications in food safety, contrasting pattern recognition with cognitive models.
• It suggests specific retail-level applications, such as automating parts of foodborne illness investigation forms.

Main Results:

• Existing AI, mainly pattern recognition, has limitations for complex retail food safety tasks.
• The development of AI based on cognitive models is proposed as a more suitable approach for advanced retail applications.

Conclusions:

• Retail food safety requires novel AI solutions beyond current pattern recognition capabilities.
• Cognitive models represent a promising paradigm shift for future AI development in food safety, enhancing capabilities for outbreak investigation and risk management.