Cornus mas L. (Cornelian cherry, CM) fruits were dehydrated by solar-drying (SD) and freeze-drying (FD), and in addition to sugar and mineral contents, the free and insoluble-bound phenolics were determined in fresh and dried fruits. After subjecting the sample to simulated in vitro digestion, the change of free and bound phenolics at gastric and intestinal digestion steps was evaluated in fresh and dried CM fruits. In fresh CM fruits, the total phenolic content (TPC) was dominated by the bound fraction, whereas the contribution of free phenolics to the total content (free + bound) became more dominant (731–1439 mg GAE/100 g dw) in the dried fruits. The bioaccessibility (BI%) of TPC from fresh CM after digestion was 193%, whereas it was 18.60 and 48.02% for SD and FD fruits, respectively. The contribution free fraction to the total TPC value was around 28% in nondigested fresh samples and increased to 94% in digested samples; however, in dried samples, it was 64% prior to digestion and only increased to 70% in digested samples. A total of 17 phenolic compounds were identified in CM fruits: chlorogenic acid, caffeic acid, epicatechin, quercetin, cyanidin-3-O-glucoside, and pelargonidin 3-O-glucoside were only detected in the free fraction; gallic acid, vanillic acid, ferulic acid, and kaempferol were detected in higher amounts in the bound fraction. The quantity of detected phenolics in the nondigested sample generally decreased from the gastric to the intestinal stage of digestion. The release of phenolics from the fruit matrix and their degradation occurred simultaneously during digestion, and this could be affected by the state of the fruit, e.g., fresh or dried. Graphical abstract

Purpose: This study aims to assess the effectiveness of different AI models in accurately aggregating information about the protein quality (PQ) content of food items using four artificial intelligence (AI) models -– ChatGPT 3.5, ChatGPT 4, Bard AI and Bing Chat.Design/methodology/approach: A total of 22 food items, curated from the Food and Agriculture Organisation (FAO) of the United Nations (UN) report, were input into each model. These items were characterised by their PQ content according to the Digestible Indispensable Amino Acid Score (DIAAS).Findings: Bing Chat was the most accurate AI assistant with a mean accuracy rate of 63.6% for all analyses, followed by ChatGPT 4 with 60.6%. ChatGPT 4 (Cohen’s kappa: 0.718, p < 0.001) and ChatGPT 3.5 (Cohen’s kappa: 0.636, p: 0.002) showed substantial agreement between baseline and 2nd analysis, whereas they showed a moderate agreement between baseline and 3rd analysis (Cohen’s kappa: 0.538, p: 0.011 for ChatGPT 4 and Cohen’s kappa: 0.455, p: 0.030 for ChatGPT 3.5).Originality/value: This study provides an initial insight into how emerging AI models assess and classify nutrient content pertinent to nutritional knowledge. Further research into the real-world implementation of AI for nutritional advice is essential as the technology develops.