AI-Driven Micro-Investment Platform for Inclusive Entrepreneurial Finance

Thesis Abstract

This study investigates how an AI-driven micro-investment platform can advance inclusive entrepreneurship by enabling small-scale founders to access early-stage funding, diversify risk, and monitor performance in real-time within emerging markets. The problem addressed is the persistent gap in capital access for micro-entrepreneurs who operate with limited collateral and opaque credit histories, which constrains startup growth and job creation. The aim is to design, implement, and evaluate a technologically grounded platform that leverages machine learning, alternative data, and behavioral insights to optimize micro-investment decision-making and participant outcomes. Specific objectives are to (i) develop an AI-enabled underwriting framework that incorporates non-traditional data sources (digital footprints, payment histories, social networks) to assess creditworthiness; (ii) implement a transparent impact-oriented investment mechanism that aligns funding with measurable entrepreneurial milestones; (iii) evaluate user adoption, trust, and financial inclusion among aspiring entrepreneurs across urban and peri-urban communities; and (iv) analyze efficiency gains in capital deployment, repayment rates, and venture survivability relative to conventional micro-finance approaches. The methodology adopts a mixed-methods, quasi-experimental design rooted in the resource-based view and signaling theory to examine capability development and lender-entrepreneur trust-building through data-driven funding. The study population comprises 2,400 micro-entrepreneurs and 15 micro-investors drawn from three metropolitan regions in a developing economy over 24 months. A stratified random sample of 1,200 entrepreneurs will be invited to participate, with 600 completing onboarding and 400 actively using the platform at multiple checkpoints. Data collection involves (i) platform log data capturing investment flows, repayment performance, and feature usage; (ii) structured surveys measuring perceived ease of use, perceived usefulness, trust, and financial inclusion indices; (iii) in-depth interviews with 60 entrepreneurs and 20 investors to explore decision rationales and risk perceptions; and (iv) official records for verification of repayment outcomes. Instrument validity will be ensured through pilot testing, content validity indexes, and Cronbach’s alpha reliability checks, supplemented by triangulation with qualitative interviews. Analytical techniques include descriptive statistics to profile the sample, logistic regression to model the likelihood of successful funding and repayment, and survival analysis to examine venture longevity. The AI underwriting component will be evaluated using receiver operating characteristic (ROC) analysis to compare predictive accuracy against traditional credit-scoring models, and techniques such as gradient boosting machines (GBM) and random forests will be employed to optimize feature selection and risk stratification. A difference-in-differences (DiD) approach will assess the impact of platform participation on business performance, while structural equation modeling (SEM) will test the relationships among perceived trust, adoption, and financial inclusion outcomes. Qualitative data will be analyzed using thematic analysis to identify patterns in trust formation, platform usability barriers, and perceived social impact, with NVivo software supporting coding consistency. Ethical considerations include informed consent, data anonymization, and compliance with data protection regulations. Expected findings include (i) improved accuracy of micro-credit decisions through integration of alternative data and AI-based risk scoring, (ii) higher rates of successful fundraising and lower default rates compared with baseline micro-finance programs, (iii) enhanced financial inclusion evidenced by increased financial literacy, savings behavior, and formal banking uptake among participants, and (iv) positive shifts in trust and perceived legitimacy of digital funding platforms among both entrepreneurs and investors. The study aims to contribute to knowledge by linking AI-enabled underwriting and transparent impact investing to inclusive entrepreneurial ecosystems, extending signaling theory in digital finance contexts, and providing a model for scalable micro-investment platforms in low-to-middle-income settings. Practical implications include guiding policy on digital financial inclusion, informing platform governance and risk management, and offering a replicable design for researchers and practitioners. The main conclusion will emphasize the viability of AI-driven micro-investment platforms as catalysts for inclusive growth, with recommendations for refining data governance, enhancing user-centric design, and expanding impact metrics to capture long-term venture viability and community benefits.