Artificial Intelligence/Machine Learning-Enabled Precision Nutrition Modestly Improves Anthropometric Outcomes and...

Precision nutrition tailors dietary guidance to individual biological, genetic, metabolic, and behavioral characteristics. Artificial intelligence (AI), machine learning (ML), wearables, sensors, mobile apps, and multiomics tools increasingly support this approach, but evidence on their effectiveness, feasibility, implementation, and equity remains fragmented. This review evaluated the effectiveness of AI/ML-enabled analytical approaches and technologies in precision nutrition across dietary, metabolic, behavioral, implementation, and equity outcomes. We conducted a Preferred Reporting Items for Systematic Reviews and Meta-Analyses-guided systematic review and meta-analysis of studies published from January 2015 to December 2025 across 6 databases. Eligible studies examined AI/ML, digital tools, wearables, mobile platforms, or multiomics approaches for personalized nutrition across the lifespan. Two reviewers independently screened studies, extracted data, and assessed bias. Random-effects meta-analyses were conducted where outcomes were comparable; remaining studies underwent qualitative and thematic synthesis. Overall, 262 studies met the inclusion criteria. Meta-analyses indicated modest pooled reductions in body weight (-1.97 kg), body mass index (BMI; -0.57 kg/m²), waist circumference (-2.41 cm), and systolic blood pressure (-4.05 mmHg), whereas no significant pooled effects were observed for lipid outcomes or glycemic markers. Corresponding 95% confidence intervals and 95% prediction intervals were -3.02 to -0.92 and -6.69 to 2.75 kg for body weight, -0.99 to -0.15 and -2.43 to 1.29 for BMI, -4.75 to -0.07 and -11.30 to 6.49 cm for waist circumference, and -6.90 to -1.19 and -13.24 to 5.15 mmHg for systolic blood pressure. Dietary intake outcomes were generally small and inconsistent, although vegetable intake showed a modest improvement. Certainty of evidence was generally low to very low because of risk-of-bias concerns, imprecision, and substantial heterogeneity; prediction intervals for several statistically significant pooled effects crossed the null. Subgroup findings were exploratory and underpowered. Technology-enabled precision nutrition may modestly improve anthropometric outcomes and systolic blood pressure, but longer, well-reported studies are needed to clarify the clinical impact of precision nutrition technologies. This trial was registered at PROSPERO as CRD420251241063.