Strategic Applications of AI and ML in US SMEs: Risk Management and Customer Service Automation

Abstract

Small and medium-sized enterprises (SMEs) are central to the U.S. economy but continue to face significant barriers to adopting artificial intelligence (AI) and machine learning (ML), particularly in risk management and customer service automation. This study examines AI/ML applications, adoption barriers, implementation frameworks, and business value creation within U.S. SMEs, with a specific focus on these two operational domains. Using a mixed-methods approach, the research integrates a systematic literature review following Okoli and Schabram (2015) with empirical analysis of data from 847 SMEs across manufacturing, retail, professional services, and technology sectors. Quantitative performance metrics are complemented by qualitative case study insights to identify adoption patterns, success factors, and implementation challenges. Findings show that SMEs adopting AI-driven risk management achieve a 67% reduction in fraud, a 43% improvement in credit risk assessment accuracy, and a 38% decline in operational risk incidents. In customer service, AIpowered chatbots handle 72% of routine inquiries, reduce response times by 84%, and lower service costs by 31%. Despite these benefits, adoption remains limited (23% for risk management and 19% for customer service), constrained by technical expertise gaps, cost concerns, data quality issues, and system integration challenges. The study contributes theoretically by extending technology adoption and dynamic capabilities frameworks to resource-constrained SME contexts. Practically, it provides actionable guidance on staged implementation, vendor selection, workforce upskilling, and change management to support effective AI/ML adoption.