Abstract: Background/Objectives: Astragalus root is a classical qi-tonifying traditional Chinese medicine that has demonstrated potential therapeutic efficacy in T2DM and NAFLD. However, the precise mechanisms underlying its effects on the comorbidity of these two disorders remain unclear. This study investigated the molecular mechanisms by which astragalus root ameliorated T2DM-NAFLD comorbidity. Methods: Network pharmacology, molecular docking, molecular dynamics simulation, and in vitro experiments were employed to elucidate the potential roles and mechanisms of astragalus root in the management of T2DM-NAFLD comorbidity. Results: A total of 25 bioactive constituents and 152 corresponding targets associated with astragalus root were identified. PPI network analysis revealed the top ten core candidate targets, among which six possessed suitable crystal structures for molecular docking, including IL-6, AKT1, JUN, TNF, CASP3, and ESR1. KEGG analysis further identified the PI3K-AKT as the most significantly en-riched pathway. Molecular docking of the principal bioactive constituent formononetin from astragalus root with the six core targets was conducted using AutoDock4 software. Molecular dynamics simulations verified the stability of the interactions between for-mononetin and each of the six core target proteins. In vitro experiments demonstrated that formononetin obviously decreased lipid droplet accumulation, downregulated TC and TG levels, suppressed the expression of TNF-α, IL-6, and IL-1β, decreased ROS and MDA levels, and enhanced GSH content and SOD activity. These therapeutical effects were achieved through inhibition of protein expression within the PI3K/AKT/mTOR signaling pathway. Conclusions: This study determined the potential therapeutic targets and underlying mechanisms of formononetin derived from astragalus root in the T2DM-NAFLD management, thereby providing a scientific basis for its clinical application.

Abstract: Climate change poses significant risks to New Zealand’s coastal agriculture through both slow-onset hazards (e.g., gradual sea-level–induced groundwater rise) and sudden-onset hazards (e.g., increasing frequency and severity of storms). These physical changes threaten the productivity and economic viability of coastal farms. However, few studies assess their combined economic impacts in a manner that supports land-use planning. This paper presents a conceptual framework to examine the implications of interacting slow- and sudden-onset climate hazards for New Zealand dairy farms, informed by real-world consultation with subject-matter experts to support feasibility analysis. We draw conclusions that illustrate the monetary impacts on farms associated with potential absorptive, adaptive, and transformational responses. The findings highlight the critical role of timing as environmental conditions deteriorate under climate change, as well as the need for policy frameworks that recognize and monetize the contribution of ecosystem services provided by coastal vegetation habitats to social, cultural, and environmental wellbeing. Incorporating these values into present-day financial decision-making is essential for supporting climate-related financial risk reduction and long-term land-use planning. Without such frameworks, the most beneficial land-use transitions are unlikely to be affordable or sustainable in New Zealand, especially towards year 2100.

Abstract: Foot-and-Mouth Disease (FMD) is a highly contagious transboundary animal disease causing substantial productivity losses and economic burdens in livestock systems. This study aimed to examine FMD transmission dynamics in Indian cattle populations and to evaluate the potential impact of key control strategies using a system dynamics approach. A Susceptible–Exposed–Infectious–Recovered–Carrier (SEIRC) model was developed to represent disease progression, incorporating causal loop diagrams and stock-and-flow structures to capture feedback mechanisms and time delays inherent to FMD epidemiology. Model simulations were conducted by systematically varying critical parameters, including the basic reproduction number (R₀), duration of infectivity, vaccination coverage, persistence of the carrier state, and duration of movement restrictions during outbreaks. The results indicated that higher R₀ values accelerated disease spread and increased peak infection levels, while shorter infectious periods resulted in more abrupt outbreaks. Vaccination coverage of at least 75% effectively stabilized the susceptible population and reduced epidemic risk. In simulated scenarios, an R₀ of 4 combined with a 14-day infectious period led to the elimination of infectious animals. Overall, the findings highlight the importance of integrated control strategies, particularly high vaccination coverage and timely movement restrictions, for reducing outbreak magnitude and duration, and provide evidence-based insights to support FMD prevention and control planning in endemic settings.

Abstract: Introduction: Obstructive sleep apnea (OSA) is a prevalent disorder in the adult population characterized by recurrent upper airway obstruction during sleep, resulting in intermittent hypoxia, sympathetic activation, and sleep fragmentation. It is linked to significant cardiovascular, metabolic, neurocognitive, and psychosocial morbidity. There is increasing evidence that continuous positive airway pressure (CPAP) adherence remains suboptimal in many patients and in those patients, surgery is often indicated. Methods: This review presents an updated, protocol-driven surgical approach grounded in clinical evidence and experience, highlighting the role of drug-induced sleep endoscopy (DISE) and personalized multi-level interventions for adult patient with OSA. Integration of anatomical phenotyping and DISE-directed planning enables precise surgical targeting. The protocol emphasizes patient selection, individualized treatment based on obstruction patterns, and perioperative optimization. This surgical algorithm improves success rates and long-term outcomes in patient’s intolerant of CPAP therapy. Results: A DISE guided, and multilevel surgical approach include: uvulopalatoplasty, septoplasty, tongue base reduction, palatoplasty and maxillomandibular advancement (MMA). Preoperative assessments include BMI and the STOP BANG along with Epworth Sleepiness scale, while postoperative care emphasizes follow up polysomnography and adjunctive therapies only when necessary. Regional experiences in Saudi Arabia and Canada underscore the importance of standardized, evidence-based surgical care. Conclusion: The purpose of this article is to establish a clear protocol for managing patients diagnosed with OSA, drawing from a review of existing literature and the insights of experienced surgeons in the field of sleep apnea, and updating current protocols with modern evidence.

Abstract: Objective: This study explores the utility of electrocardiogram parameters in conjunction with machine learning models for the early diagnosis of neonatal transient tachypnea (TTN). TTN is a common cause of respiratory distress in neonatal intensive care units, and early diagnosis has the potential to reduce invasive interventions and shorten hospital stays. Methods: The study retrospectively examined data from 101 neonates diagnosed with TTN and 82 healthy neonates, utilizing parameters such as P, QRS, T angles, and frontal QRS-T angle obtained from ECG. Results: Decision Tree, Neural Network, Random Forest, Boosting, and Support Vector Machine models were utilized among the machine learning algorithms. The dataset was split into 65% for training, 20% for validation, and 15% for testing. According to the findings, the Random Forest classification model demonstrated superior performance compared to other models, achieving 71.4% test accuracy, an average AUC value of 0.790, and a Matthews Correlation Coefficient of 0.443. The MCC value indicated that the Random Forest model possesses reliable predictive power even with imbalanced datasets. Notably, ECG parameters such as PR interval, V2 T voltage, and SV1 voltage were identified as the most significant features influencing the model's predictive performance. Conclusions: These findings suggest that ECG-based machine learning models can enhance clinical decision-making by facilitating non-invasive, rapid, and accurate diagnosis of TTN. Such artificial intelligence-driven systems hold the potential to mitigate unnecessary interventions, expedite treatment initiation, and improve neonatal prognoses. Future efforts should focus on enhancing model interpretability through the incorporation of explainable AI methodologies to facilitate their seamless integration into clinical practice.

Abstract: Objective: The development of ABCB1-mediated resistance limits the clinical efficacy of paclitaxel. Lapatinib is a small-molecule reversible tyrosine kinase and ABCB1 inhibitor that could prevent resistance. Our objective was to de-termine a recommended phase 2 dose (RP2D) of the combination of paclitax-el and lapatinib. Methods: A phase 1 dose-escalation study utilizing a Bayesian optimal interval (BOIN) design in recurrent ovarian cancer patients. Patients were pretreated with pulsed lapatinib in the 48 hours preceding weekly paclitaxel (80 mg/m2) in 28-day cycles for up to 3 cycles. We evaluated three lapatinib doses, escalating from 750 to 2,000 mg orally twice daily. Results: Sixteen patients were eligible and evaluable for efficacy and toxicity. Patients received a median of three prior therapies. Three patients were treated at dose level 1, six at dose level 2, and seven at dose level 3. There was one dose-limiting toxicity (DLT) in dose level 2 (diarrhea) and another in dose level 3 (neutropenia), with a posterior DLT estimate of 0.17, 95% credible in-terval of (0.01, 0.53) for dose level 3 based on isotonic regression. The most common grade 1-2 adverse effects were diarrhea (87.5%), leukopenia (56.3%), and anemia (50%). One (6.25%) patient had a complete response, and 7 (43.75%) patients had partial responses for an overall response rate (ORR) of 50%. The clinical responses are supported by a significant decreasing trend in CA 125 over six cycles (p=0.0001). Among the seven patients treated at the RP2D, the ORR was 71.4%. Conclusions: The combination of paclitaxel and lapatinib is safe and with an efficacy signal. The RP2D is weekly paclitaxel 80 mg/m2 combined with lapatinib 2,000 mg twice daily two days before the paclitaxel dose. This trial was registered at ClinicalTrials.gov ID: NCT04608409.

Abstract: Research on forecasting corporate financial performance has rushed from traditional econometric models toward machine learning, deep learning, and high-precision hybrid AI architectures. These methods can capture nonlinear relationships, high-dimensional structures, and regime shifts in financial data more effectively, which has driven their widespread adoption. At the same time, practical requirements for interpretability, regulatory transparency, and model risk governance have made explainable AI an essential component of modern forecasting systems. This Structured Literature Review synthesizes ninety-three empirical studies published between 2000 and 2025 using a PRISMA-informed selection procedure. It evaluates the actual contributions of hybrid AI and explainable AI to corporate financial performance forecasting. The review shows that econometric and machine learning hybrids, ensemble learning models, DEA-based machine learning frameworks, deep learning combined with signal processing, and multimodal architectures are extensively used and collectively improve predictive accuracy and stability. Methods such as SHAP, LIME, partial dependence, and individual conditional effect analyses, attention mechanisms, and counterfactual reasoning significantly enhance model interpretability, support managerial decision-making, and strengthen compliance with regulatory expectations. Despite these advances, challenges remain, including the predominance of static data analysis, limited generalizability, and the lack of architectures designed for realistic deployment. Future research should focus on multimodal data integration, causal AI, adaptive, real-time learning frameworks, and explainable hybrid systems aligned with regulatory and governance requirements.

Abstract: We provide generalizations of the conventional logistic population dynamics models suitable for periodic breeders, such as migratory birds, occupying varied habitats during the breeding cycle. These models require separate density dependencies for the birth and death rates, which may be habitat specific. Some analytical functional forms for the density dependencies are discussed where the population controlling mechanisms are each characterized by a distinct carrying capacity and saturation power. Multiple mechanisms might be operative simultaneously with the smallest carrying capacity usually dominating, but subject to influence from the others. We compare the dynamics and applicability for corresponding continuous differential and discrete difference population models. Generally, the differential models are stable, but exhibit repetitive seasonal variations for periodic breeders. The inherent delays in the discrete models may yield instabilities for large birth rates, as is known for single habitats, and may lead to significant discrepancies from the differential models for periodic breeders. The discrete models are also applicable to the life cycles of metamorphic and spawning species with non-overlapping generations. Threshold effects are also considered.

Abstract: Background: Healthcare professionals experience continuous biological and psychosocial stressors that may disturb oral and systemic homeostasis. Alterations in salivary secre-tion, mucosal immunity, and microbiome composition reflect adaptive cellular responses to chronic occupational stress. Understanding these mechanisms may provide a biologi-cal framework for resilience and wellbeing in clinical everyday practice. Objective: To narratively review the evidence linking oral cellular and molecular mechanisms, -salivary biomarkers, epithelial and immune cell activity, and microbiome dynamics- with stress, fatigue, burnout, and wellbeing outcomes among healthcare professionals. Methods: A PRISMA-guided search of PubMed, Scopus, Web of Science, and Cochrane Oral Health identified studies investigating oral cellular or molecular parameters in relation to occu-pational stress or wellbeing indicators in healthcare settings. Eligible designs included observational, experimental, and interventional studies. Data were extracted using stand-ardized forms, quality was appraised via ROBINS-I and the Newcastle-Ottawa scale, and results were synthesized thematically. Results: Evidence from 99 studies suggests that chronic occupational stress elevates salivary cortisol, oxidative stress markers, and pro-inflammatory cytokines (IL-6, TNF-α), while reducing protective salivary immuno-globulin A and microbiome diversity. Balanced oral immune and microbial profiles were associated with better psychological adaptation and lower fatigue indices. Conclusions: Oral cellular homeostasis offers a promising window into the biological underpinnings of occupational stress and resilience in healthcare professionals. Systematic integration of salivary and mucosal biomarkers into workplace wellbeing programs could enhance ear-ly detection of dysregulated stress physiology. Future interdisciplinary research should bridge oral biology, occupational medicine, and mental health to strengthen sustainable wellbeing strategies across the health workforce.

Abstract: Substrates have become essential enabling materials for creating lightweight electronic components, particularly supporting advanced telecommunication technologies. This progress is driven by continuous advancements in novel substrate materials and cut-ting-edge fabrication techniques, pushing the limits of high-frequency device design. This paper explores both the challenges and breakthroughs in 5G mmWave substrate technology, focusing on recent developments in materials, device fabrication and integration methods that enhance performance and providing an in-depth analysis on the importance of mmWave technology. This paper highlights the key concerns in substrates design to researchers and academicians accelerates invention and commercialization of substrate designs in areas such as antenna engineering and integrated circuit technologies as well as addressing key issues like scalability and thermal impact in flexible substrates. Since matters related to material losses and substrates’ fabrication constraints are increasingly severe at high frequencies, mmWave substrates are highly needed to be look at, therefore this paper details the particular issues related to mmWave propagation and manufacturing design processes for high-frequency devices. Aims at optimizing antenna and system reliability by employing advanced design and materials as well as outlines the existing gaps that need a clarification to augment 5G mmWave infrastructure and services.

Abstract: The study aims to investigate the effect of full-fat rice bran (FFRB; Tainung No. 81, Taiwan) at various doses on insulin resistance, muscle atrophy, and gut microbiome in ovariectomized (OVX) mice fed a high-fat diet (HFD). Thirty-six female ICR mice were grouped into six: young sham-operated mice fed an AIN-93M diet, OVX mice fed an AIN-93M diet, OVX mice fed a HFD, and OVX mice fed a HFD with 5%, 10%, and 20% FFRB. FFRB intervention attenuated HFD-induced weight gain and visceral fat accumulation, improved insulin resistance, and enhanced grip strength in OVX mice. Notably, 20% FFRB significantly upregulated muscle protein synthesis genes (MyoG, mTOR, eIF-4EBP1) and downregulated muscle atrophy markers (FOXO1, MuRF-1), while reducing the inflammatory cytokine IL-6. 20% FFRB also improved gut barrier integrity by upregulating colonic tight junction genes (Occludin and ZO-1) and increased the abundance of SCFAs-producing bacteria Muribaculum genus. 10% FFRB significantly downregulated FOXO1 and increased the abundance of Lachnospiraceae_UCG-001 genus. In conclusion, FFRB intervention, particularly at 20%, effectively mitigated HFD-induced insulin resistance and muscle atrophy, potentially through modulation of gut microbiota and enhancement of gut barrier function.

Abstract: Virtual Reality (VR) has emerged as a powerful medium for immersive human–computer interaction, where users’ emotional and experiential states play a pivotal role in shaping engagement and perception. However, existing affective computing approaches often model emotion recognition and immersion estimation as independent problems, overlooking their intrinsic coupling and the structured relationships underlying multimodal physiological signals. In this work, we propose a modality-aware multi-task learning framework that jointly models emotion recognition and immersion estimation from a graph-structured and symmetry-aware interaction perspective. Specifically, heterogeneous physiological and behavioral modalities—including eye-tracking, electrocardiogram (ECG), and galvanic skin response (GSR)—are treated as relational components with balanced structural roles, while their cross-modality dependencies are adaptively aggregated to preserve interaction symmetry and allow controlled asymmetry across tasks, without introducing explicit graph neural network architectures. To support reproducible evaluation, the VREED dataset is further extended with quantitative immersion annotations derived from presence-related self-reports via weighted aggregation and factor analysis. Extensive experiments demonstrate that the proposed framework consistently outperforms recurrent, convolutional, and Transformer-based baselines, achieving higher accuracy (75.42%), F1-score (74.19%), and Cohen’s Kappa (0.66) for emotion recognition, as well as superior regression performance for immersion estimation (RMSE = 0.96, MAE = 0.74, R-squared = 0.63). Beyond empirical improvements, this study provides a structured interpretation of multimodal affective modeling that highlights symmetry, coupling, and symmetry breaking in multi-task learning, offering a principled foundation for adaptive VR systems, emotion-driven personalization, and dynamic user experience optimization.

Abstract: We analyzed the behavior of a “Solar Wall” and validated the apparatus with three nanofluids (silver, titanium dioxide and a hybrid compound) in view of their photothermal conversion performance. The factors considered were the temperature gain in relation to the base fluid, the stored energy and the specific absorption rate. A cost survey was carried out to find out the profit of each nanofluid. Five concentrations were studied for each nanofluid. A hybrid nanofluid formed by the previous ones was also tested. The results presented that the Solar Wall has achieved repeatability, and we can state that it is suitable for the tests on other nanofluids. The silver and hybrid nanofluids performed better, the first obtained a temperature gain of 10.2 °C compared to the base fluid and the hybrid reached 9.9 °C. Regarding the energy gain, the silver-based obtained a gain of 31.93%, and the hybrid obtained 34.52%. The SAR values for the silver nanofluid were higher than the titanium-based, nevertheless the cost to generate an energy unit using the former was higher than in the titanium case. The silver-based and the hybrid nanofluids obtained improved photothermal conversion, being the most promising options.

Abstract: Existing patent examination approaches face fundamental limitations: they struggle with comprehensive prior art coverage due to maximum similarity scoring without considering all claim elements, provide limited ranked retrieval through binary classification without confidence scoring, and incur substantial computational overhead while generating generic outputs that miss claim-specific details. To address these challenges, we introduce \textbf{Integrated Patent Prior Art Search with Claim-Aware Retrieval and Novelty Assessment} (IPAS-CARNA), a novel three-stage pipeline combining enhanced claim-document matching, continuous novelty assessment, and claim-aware summarization. Our approach models element-wise claim coverage through adaptive chunking and weighted aggregation, integrates continuous novelty scoring with confidence assessment, and introduces claim-aware summarization with dynamic length control. Extensive experiments on CLEF-IP 2013, USPTO examination records, and HUPD validation sets demonstrate significant improvements: MAP@100 of 0.342 with 14.8\% improvement in retrieval recall, 18.2\% improvement in NDCG@10 for novelty ranking, technical accuracy above 0.85, and ROUGE-L scores of 0.456 for summarization. Our work establishes an effective integrated solution for automated patent prior art analysis.

Abstract: For a long time, the "biogenic theory" of petroleum origin has dominated mainstream thinking, positing that petroleum forms from the burial and thermal evolution of ancient microbial, plant, and animal remains in sedimentary environments. However, traditional theories cannot fully explain how complex biological macromolecules precisely crack into relatively simple hydrocarbon small molecules over geological time scales, and controversies remain regarding the energy sources and kinetic mechanisms of the cracking process. Integrating the core physical principle that cosmic expansion induces atomic expansion[1][3], we can construct a novel petroleum generation mechanism: petroleum is a product of sedimentary organic matter derived from microbial, plant, and animal remains, which undergoes gradual cracking and recombination of molecular structures under the sustained action of atomic expansion driven by cosmic expansion over hundreds of millions of years, ultimately transforming from complex macromolecules into hydrocarbon small molecules.

Abstract: Allosteric modulation has emerged as a powerful strategy for achieving superior selectivity and safety in drug discovery and protein function regulation. Unlike highly conserved orthosteric sites, allosteric pockets are structurally diverse and less evolutionarily constrained, making them particularly amenable to be modulated by designed miniproteins. Miniproteins can provide extended binding interfaces and high affinity for shallow, dynamic, or cryptic regulatory surfaces that are often inaccessible to small molecules. Recent advances in artificial intelligence (AI) are transforming this field through deep learning–based structure prediction and generative modeling. These AI-driven approaches enable the identification of allosteric hotspots, characterization of conformational ensembles, and de novo} design of structured miniprotein binders. They are rapidly expanding the landscape of designing selective modulators across diverse allosteric targets, including GPCRs, receptor tyrosine kinases, nuclear receptors, ion channels, and protein–protein interfaces. This review summarizes state-of-the-art AI-driven computational methodologies for designing miniproteins as potential allosteric modulators and discusses their current challenges and future opportunities in allosteric drug discovery.

Abstract: Although the number of Islamic mutual funds (IMFs) in Indonesia has grown in recent years, the market remains relatively small compared with countries such as Malaysia and Saudi Arabia. This study explores the key drivers and barriers influencing IMF development and Indonesian investors’ attitudes toward these products. Drawing on 21 semi-structured interviews with investment managers, regulators, and policymakers, the study identifies major constraints, including regulatory complexity, a limited and illiquid pool of Shariah-compliant securities, suboptimal fund performance, tax limitations, low product awareness, limited Islamic financial literacy, and moderate investor risk orientation. Simultaneously, several factors support potential growth, such as robust investor protection regulations, a large and youthful population, opportunities for product innovation, and strong government backing. The findings offer practical guidance for enhancing product offerings, improving the regulatory framework, and promoting financial literacy, while underscoring that fund performance, investor awareness, and financial literacy remain central determinants of investment behavior.

Abstract: We propose a causal-diamond formulation of semiclassical gravity in which a finite-resolution boundary regulator (Coherency Screen) supplies the minimal edge structure required for a local description in a Wheeler–DeWitt setting. Diamond-local dynamics are defined by an informational variational principle: for each diamond O, the effective cost functional is the relative entropy S_rel(ρ_O || σ_O[g]) between the reduced physical state and a geometric reference family. In the small-diamond modular/KMS regime, a derivative expansion of this cost, implemented via a heat-kernel spectral expansion, yields a local effective action whose leading terms recover the Einstein sector and select a spinorial (Dirac-type) transport structure. A discrete edge-mode counting, together with Newton’s constant G, fixes a characteristic resolution scale M_s ~ 3×10^13 GeV. Treating M_s as the onset of the leading stiffness correction places the high-curvature regime in a plateau universality class, giving a capacity-set scalar amplitude and a tensor target r ~ 10^-3. We further discuss how the same boundary logic constrains the gauge and mass sectors in a spectral-action-compatible formulation, suggesting discrete relations among effective coupling normalizations and a structured organization of charged-lepton scales via geometric accessibility of the boundary algebra. We also outline late-time phenomenological extensions in which finite-resolution boundaries induce a mild running of effective stiffness and horizon-set acceleration scales. Overall, the construction yields a compact set of correlated, falsifiable targets tied to a single microscopic resolution scale.

Abstract: We develop a symmetry-based reconstruction of the vacuum impedance and the fine-structure constant. Hyperbolic geometry and discrete sectorization of the electromagnetic field plane are the only input assumptions. The construction identifies a unique integer-square hyperbolic selector that fixes the electric–magnetic partition without adjustable parameters. This yield the geometric part of the vacuum impedance when combined with the quantum scale $h/e^{2}$. The same discrete structure provides a normalization for the fine-structure constant through a universal sector angle $\pi/24$, connecting topological quantization phenomena in metals and alloys, including Berry phases, Zak phases, and quantized Hall responses. The resulting framework places electromagnetic constants within a unified geometric–topological setting and suggests experimentally accessible consequences in systems with discrete rotational or modular symmetry.

Abstract: This study examines how artificial intelligence (AI) contributes to contemporary processes of authenticity evaluation by functioning as a multimodal diagnostic cue in consumer decision-making. Drawing on survey data collected from 468 visitors at Terra Madre Salone del Gusto in Turin, Italy, the study tests a structural model comprising five latent constructs: Authenticity Trust, Perceived AI Usefulness and Diagnosticity, Multimodal Value, User Engagement, and Behavioural Intentions. The findings indicate that heritage-based and institutional authenticity cues remain foundational in consumers’ evaluations, but are increasingly interpreted and conditionally reinforced through interaction with AI-mediated information perceived as credible and diagnostically informative. Multimodal inputs—particularly the integration of textual, visual, and auditory narratives—are associated with richer authenticity perceptions and higher levels of user engagement. Experiential enjoyment during interaction with the AI system is positively related to intentions to adopt AI-supported evaluation tools, while behavioural intentions also encompass a willingness to pay a premium for products confirmed as authentic. Although the use of a convenience sample limits generalisability, the results highlight the broader potential of multimodal AI systems to reduce evaluative uncertainty and support trust formation in complex cultural and consumer environments. Conceptually, the study advances the notion of augmented authenticity, defined as a hybrid evaluative process in which tradition-based trust mechanisms are dynamically interpreted and reinforced through perceived AI diagnosticity and multimodal coherence. By situating AI within culturally embedded processes of meaning-making rather than purely instrumental evaluation, the findings contribute to interdisciplinary debates on technology-mediated trust, consumer judgement, and the societal implications of AI-assisted decision-making.