The rapid growth of wealth-technology (wealth-tech) platforms has intensified the importance of digital trust, particularly among Generation Z investors who rely heavily on social media–driven information sources when making investment-related decisions. While prior studies have examined influencer marketing, electronic word-of-mouth (e-WOM), and social media engagement in fintech contexts, empirical research that integrates these persuasion mechanisms into a unified trust-based model of wealth-tech adoption intention remains limited. Drawing on Source Credibility Theory, Trust Transfer Theory, and digital engagement frameworks, this study proposes and tests an integrative model in which influencer credibility, e-WOM, and social media engagement simultaneously influence wealth-tech adoption Intention through the mediating role of digital trust. Using survey data collected from 255 Generation Z actual users of wealth-tech platforms in Indonesia, this study employs Structural Equation Modeling (SEM) to simultaneously test the measurement model and the proposed trust-based structural relationships, including the mediating role of digital trust. A purposive sampling approach was adopted to ensure respondents possessed direct experience with wealth-tech applications, thereby enhancing construct validity in this specialized digital investment context. The results indicate that influencer credibility, e-WOM, and social media engagement each exert a significant positive effect on digital trust, which in turn strongly influences wealth-tech adoption Intention. Digital trust is found to play a critical mediating role, reinforcing its central importance in investment-oriented digital platforms characterized by heightened perceived risk. This study contributes to the literature by extending digital trust and fintech adoption research in three ways: (1) by integrating multiple digital persuasion mechanisms into a single trust-centered framework, (2) by empirically validating digital trust as a key mediating mechanism in a wealth-tech investment context, and (3) by providing contextual insights from an emerging market characterized by rapid digital adoption and persistent trust challenges. Practically, the findings offer guidance for wealth-tech platforms and digital marketers in designing trust-enhancing strategies targeting Generation Z investors.

Endometriosis remains underdiagnosed due to reliance on invasive laparoscopy. Artificial Intelligence (AI) using either imaging or structured clinical data have shown promise, but single modality approaches face limitations in sensitivity, calibration, and clinical reliability. This work seeks to evaluate whether decision-level multimodal fusion of Magnetic Resonance Imaging (MRI)-based and clinical data-based AI systems improves diagnostic performance, calibration, and net clinical benefit, compared with single-modality models. Two previously validated models were combined with retrospective data from 1,208 patients with suspected endometriosis: a Dual U-Net trained on pelvic MRI with Gradient-weighted Class Activation Mapping (Grad-CAM) interpretability and a dense neural network trained on structured clinical features with SHapley Additive exPlanations (SHAP). This study tested weighted averaging, stacking via logistic regression, and confidence-gating. Performance was assessed using accuracy, precision, recall, F1-score, and area under the curve (AUC). Calibration was evaluated using the Brier score, expected calibration error (ECE), and reliability diagrams. Clinical utility was quantified with decision curve analysis (DCA). Statistical significance was tested with McNemar’s test for accuracy and DeLong’s test for AUC. Multimodal fusion outperformed both single modality models. Weighted averaging accuracy was 0.89, precision was 0.89, recall was 0.87, and F1-score was 0.86, thus improving on either modality alone. Stacking further enhanced calibration (ECE reduction from 0.8 to 0.04) and yielded higher net benefit across clinically relevant probability thresholds (20 to 60%). DCA indicated fusion would avoid 12 to 18 unnecessary surgical investigations per 100 patients, compared with single modality strategies. Confidence-gating maintained performance under simulated distribution shifts to support robustness. Decision-level multimodal fusion enhanced non-invasive diagnosis of endometriosis by improving accuracy, calibration, and clinical utility. These results demonstrated the value of integrative AI gynecological care and justify prospective validation in real-world clinical settings.

This work provides a complete methodology for adopting well-established AI methods (predictive analytics, LLM agents, forecasting) into Microsoft Dynamics 365 Customer Relationship Management (CRM) for agricultural lending. While not claiming that the algorithms are novel, this work contributes a pragmatic approach to implementing these algorithms that specifically address the regulatory, seasonal, and operational characteristics of agricultural finance, as regulated by the Farm Credit System. It focuses on the real-life constraints and constraints within the regulated financial services industry, and measurable impacts that occurred. The paper provides a domain-oriented application of specific existing AI-CRM integration, with credible statistical testing including an external validation on USDA datasets and benchmarking across peer Farm Credit institutions, as well as cross-institutional analysis. By taking a reasonably conservative duration of 18 months, the Farm Credit institutions noted a statistically significant impact (operational efficiencies of the lending institution to assess member interests) where average case resolution time reduced by 28% (67.2h to 48.4h), and lead conversions improved by 35% (25.9% to 35.0%). Each methodology of implementation also included a series of validations in compliance with regulatory oversight in financial institutions that started to build data governance, model performance compliance through a proactive risk definition, and compliance standards suitable for their institution, and within regulatory standards by regulations. Beyond statistical significance (paired tests, $p <0.001$), practical impact was quantified using absolute and relative changes and bootstrap confidence intervals. The article provides the agricultural lending industry an applied methodology to adopt AI for stakeholder innovation while ensuring they are adept in their enterprise risk management requirement, and still target measurable business outcomes. Given a conservative potential implementation timetable (i.e., 18 months) and validation methodology protocols developed to ensure complete data and model validation, this approach is scalable for agricultural lending implementation and would be a useful instrument across all 72 Farm Credit System institutions.