Generative artificial intelligence ChatGPT in clinical nutrition — Advances and challenges

ChatGPT and other artificial intelligence (AI) tools can modify nutritional management in clinical settings. These technologies, based on machine learning and deep learning, enable the identification of risks, the proposal of personalized interventions, and the monitoring of patient progress using data extracted from clinical records. ChatGPT excels in areas such as nutritional assessment by calculating caloric needs and suggesting nutrient-rich foods, and in diagnosis, by identifying nutritional issues with technical terminology. In interventions, it offers dietary and educational strategies but lacks critical abilities such as interpreting non-verbal cues or performing physical examinations. Recent studies indicate that ChatGPT achieves high accuracy in questions related to clinical guidelines but shows deficiencies in integrating multiple medical conditions or ensuring the accuracy of meal plans. Additionally, generated plans may exhibit significant caloric deviations and imbalances in micronutrients such as vitamin D and B12. Despite its limitations, this AI has the potential to complement clinical practice by improving accessibility and personalization in nutritional care. However, its effective implementation requires professional supervision, integration with existing healthcare systems, and constant updates to its databases. In conclusion, while it does not replace nutrition experts, ChatGPT can serve as a valuable tool to optimize nutrition education and management of our patiens, always under the guidance of trained professionals.