Can artificial intelligence uncover the bioactive peptides’ benefits for human health and knowledge? A narrative review

The intersection of Artificial Intelligence (AI) and food science has opened new frontiers in understanding the "dark matter" of food, the vast array of unidentified bioactive compounds that influence human health. This narrative review examines how AI, particularly machine learning and deep learning, is revolutionizing the discovery, characterization, and application of bioactive peptides and amino acids derived from food sources, both plant- and animal-based. These compounds exhibit diverse health benefits, including antioxidant, anti-inflammatory, antihypertensive, and antimicrobial properties, yet their complexity and the limitations of traditional methods have hindered comprehensive study. AI-driven approaches, such as predictive modeling, molecular dynamics simulations, and natural language processing, are accelerating the identification of bioactive peptides, optimizing extraction processes, and enabling personalized nutrition strategies. The integration of AI with omics technologies (e.g., nutrigenomics, proteomics) further enhances our understanding of how these peptides modulate physiological pathways. However, this is not without challenges and limitations, such as data quality, model interpretability, and persistent gaps in interdisciplinary collaboration. Additionally, the review highlights the lack of standardized databases and concerns about the use of AI, including the need for ethical approvals and protocols aligned with privacy laws, particularly in the context of personalized nutrition guidance. This review synthesizes current advancements, identifies research gaps, and underscores the transformative potential of AI in functional food development and precision nutrition. By addressing these challenges, AI can unlock the full therapeutic potential of food-derived bioactive compounds, providing innovative solutions to global health challenges such as non-communicable diseases. The findings advocate for robust interdisciplinary efforts to bridge computational and nutritional sciences, paving the way for scalable, evidence-based applications in health and wellness.