From Algorithms to Capabilities: A Cross-National Analysis of Digital Technology's Impact on Financial Literacy with Equity Implications

Abstract

This study investigates the transformative potential and equity implications of digital intelligence technologies (DIT) in financial literacy education through a cross-national analysis of PISA 2022 data (N = 98,000 students, 20 countries/regions) and a quasi-experimental intervention in China. Employing multilevel linear regression, structural equation modeling, and difference-in-differences methodologies, we establish that DIT significantly enhances financial literacy (β = 0.756 SD, p < 0.001) through four interdependent mechanisms: technical capability (β = 0.342), learning behaviors (β = 0.287), security awareness (β = 0.198), and innovative application (β = 0.156). However, substantial heterogeneity reveals critical disparities: urban students gained 61.3% more than rural peers, while high-socioeconomic status (SES) students outperformed low-SES counterparts by 112.7%. These disparities were further amplified by institutional stratification, with resource-advantaged "key schools" increasing benefits by 58.9%. China's integrated intervention model, which combines teacher training, cloud-based VR labs, and algorithmic audits — demonstrated an 18.65-point net gain, proving that ecosystem design moderates DIT efficacy (school infrastructure β = 0.092; family digital capital β = 0.078). We conclude that while DIT advances financial capabilities, its equitable deployment requires targeted infrastructure investment, teacher re-skilling, and ethical safeguards against algorithmic bias. This research contributes a validated theoretical framework for DIT-empowered financial education and provides evidence-based pathways for human-AI collaboration in financially digitized societies.