Background: Microbial contamination in the global food industry, driven by the increasing foodborne illness and food spoilage, brought the antimicrobial bioactive compounds into focus. The conventional screening methods are time-consuming, labour-intensive, and costly. Artificial intelligence (AI) and machine learning (ML) algorithms can efficiently screen top-performance candidates, appearing as transformative tools in the discovery of antimicrobials.

Scope and approach: We assess traditional methods for screening antimicrobial agents, categorizing them according to the diffusion pathways of bioactive compounds. It also explores the integration of AI and ML technologies in the food field, highlighting advancements in algorithms, improvements in databases, and the expansion of computing resources. Additionally, this review delves into examples of AI-predicted antimicrobial compounds, also discussing their validation and testing processes as promising applications in food systems.

Key findings and conclusions: Conventional methods have limitations including the need for extensive testing, while AI-driven screening technologies provide rapid and efficient identification of a large number of potentially bioactive candidate compounds. Despite facing challenges in quality, quantity, annotation, and web-accessibility of databases, AI, and ML-based technologies hold potential for screening antimicrobial peptides for food applications. A future direction of the field includes the expansion of antimicrobial bioactive compounds databases to include a wider variety of sources, incorporating high-quality - annotations. Culminating in personalized recommendations for optimizing antimicrobial usage would be achieved by integrating multi-omics data, optimizing the structure of commercial antimicrobials, and developing decision support systems.