Small businesses in underserved US communities that report higher AI adoption also report substantially better financial literacy, decision-making and higher profit growth (9.5% vs 5.8%), but gains are curtailed by low training uptake, uneven education and demographic barriers.

Main Finding

AI adoption among small businesses in underserved U.S. communities is positively associated with higher financial literacy, better decision-making quality, and faster profit growth — but these benefits are conditional. Firms with high AI adoption reported significantly higher financial literacy and decision-making scores and an average profit growth of 9.5% versus 5.8% for low adopters. Adoption and impact are constrained by low AI/digital training participation, uneven education, and demographic disparities (gender, age). The transformative potential of AI therefore depends on complementary human capital, contextualized tools, and supportive ecosystems.

Key Points

• Theoretical framing: Resource-Based View (RBV) — AI treated as a strategic intangible resource whose value depends on complementary capabilities (data, skills, routines).
• Sample and scope: Quantitative survey of 400 small business owners in underserved U.S. communities.
• Outcomes measured: AI adoption level, financial literacy, decision-making quality, and business growth (profit growth rates).
• Core empirical result: Positive, statistically significant correlation between AI adoption and (i) financial literacy, (ii) decision-making quality, and (iii) profit growth (9.5% for high adopters vs. 5.8% for low adopters).
• Mediating barriers identified:
  • Low participation in AI/digital training programs.
  • Uneven educational backgrounds and digital literacy.
  • Demographic disparities (notably by gender and age).
  • Infrastructure, trust, and localized relevance gaps (discussed in literature review).
• Risks highlighted: Without equitable design and oversight, AI can reproduce or amplify exclusion (algorithmic bias favoring formal-income profiles; linguistic and cultural mismatch).
• Recommended orientation: Bundle technological deployment with digital upskilling, inclusive design, governance, and multi-stakeholder collaboration.

Data & Methods

• Methodology: Quantitative cross-sectional survey using a structured questionnaire.
• Sample: 400 small business owners operating in underserved communities in the USA.
• Key measures (as reported): self-reported AI adoption level, standardized/constructed financial literacy scores, assessed decision-making quality, and reported profit growth rates.
• Analysis (reported findings): correlation and significance testing showing higher scores and growth among high AI adopters; mediation/constraint analysis highlighting training, education, and demographic factors as limiting effective adoption.
• Limitations (explicit or implied by the paper):
  • Cross-sectional design — limits causal inference and long-term impact assessment.
  • Reliance on self-reported measures for some outcomes (potential measurement bias).
  • Limited detail on sampling frame, questionnaire instruments, and specific statistical models in the reported excerpt (full paper may supply these).

Implications for AI Economics

• AI as a firm-level economic asset: The paper supports treating AI as a VRIN-like intangible under RBV — its economic value emerges through complementarities (human capital, firm data, routines), not from technology alone.
• Distributional effects: AI can increase productivity and firm performance but risks widening inequalities without targeted interventions; economists should model adoption heterogeneity and complementary investments when estimating aggregate gains from AI in the SME sector.
• Policy and market interventions:
  • Invest in digital and AI literacy targeted to underserved entrepreneurs to unlock value from AI tools.
  • Subsidize or support contextualized AI solutions (language, cultural relevance, local data) and low-cost access to infrastructure.
  • Design incentive structures (grants, matched funding, technical assistance) that encourage bundling AI deployment with training and governance mechanisms.
  • Regulate and audit AI-driven credit/decision tools to detect and mitigate algorithmic bias; require transparency and use of alternative data that fairly represent informal incomes.
• Research directions for AI economics:
  • Longitudinal and experimental studies to identify causal effects of AI-enabled interventions on SME survival, revenue growth, and financial resilience.
  • Cost–benefit analyses comparing pure technology subsidies versus bundled interventions (technology + training + governance).
  • Models of equilibrium effects: how differential AI adoption across communities affects local markets, credit access, and inequality.
  • Quantitative assessment of algorithmic bias in SME-focused AI products and the economic welfare consequences for marginalized groups.
• Practical takeaway for economists and policymakers: Estimating the economic returns to AI in small-business contexts requires accounting for complementarities (skills, trust, infrastructure) and distributional constraints; policy that supports those complementarities can materially increase the inclusive economic gains from AI.