Generative AI in Precision Nutrition: A Review of Current Developments and Future Directions.

Area of Science:

• Nutritional Science
• Artificial Intelligence
• Biotechnology

Background:

• Precision nutrition (PN) personalizes dietary guidance using individual variability.
• Traditional AI advanced nutritional research but lacked adaptive capabilities.
• Generative AI (GenAI) offers adaptive interventions for enhanced PN, but its scope is unclear.

Purpose of the Study:

• To review original works applying GenAI in PN.
• To focus on GenAI applications, methodologies, and limitations in PN.
• To characterize the current state of GenAI in personalized nutrition.

Main Methods:

• Systematic literature search across PubMed, ACM Digital Library, and Scopus.
• Inclusion of original works deploying GenAI models in PN.
• Formal assessment of included works on data, validation, transparency, bias, and security.

Main Results:

• 21 eligible studies published after 2024 were identified.
• LLM-based systems for personalized dietary recommendations are surging.
• Limitations include synthetic data evaluation, hallucinations, and minimal biological/genomic integration.

Conclusions:

• GenAI in PN is rapidly expanding but often lacks biological determinants.
• Personalization is primarily based on user preferences, not biological factors.
• Standardized reporting, genome-informed modeling, and human-in-the-loop validation are crucial for holistic PN.