Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2126
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-21T22:02:39.597232Z
status: published
lens: g_space
sector: crime
headline: Financial self efficacy as a mediator of artificial intelligence use in detecting money laundering and financial fraud in Jordanian Islamic banks
dek: This study examines the mediating role of financial self-efficacy in the relationship between artificial intelligence (AI) adoption and the perceived effectiveness of detecting money laundering and financial fraud in Jordanian Islamic banks. The study addresses a behavioral and institutional gap in the literature by explaining how AI-supported monitoring capabilities may be translated into effective regulatory action through employees’ confidence in interpreting and using AI-generated outputs. A quantitative cro…
gain_title: Among 115 compliance and audit staff in Jordanian Islamic banks, higher AI adoption was associated with greater perceived AML and fraud detection effectiveness, partly through employees' financial self-efficacy.
problem_title: (none)
trace_subject: (none)
gain_reading: Among 115 compliance and audit staff in Jordanian Islamic banks, higher AI adoption was associated with greater perceived AML and fraud detection effectiveness, partly through employees' financial self-efficacy.
gain_evidence: AI adoption has a statistically significant positive direct effect on perceived AML/fraud detection effectiveness | The mediation analysis shows a significant indirect effect of AI adoption on perceived detection effectiveness through financial self-efficacy
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers surveyed 115 employees in compliance, Shariah supervision, internal audit, auditing, and risk management roles at Jordanian Islamic banks. They tested whether employees' confidence in using AI outputs links AI adoption to how effective they judge money laundering and fraud detection to be.

The results suggest that AI tools matter partly through the people who interpret them, so training and oversight may shape whether monitoring translates into regulatory action. Because the data are perceptions from a single cross-sectional survey, the findings do not show actual detection rates or causation.
limitation: The study measures perceived detection effectiveness from 115 employees using a cross-sectional survey, so it does not establish causal effects or actual detection outcomes.
tag: Evidence-backed gain
key_points: The study used a quantitative cross-sectional questionnaire with 35 items covering AI adoption, financial self-efficacy, perceived detection effectiveness, and demographics. | AI adoption had a positive effect on financial self-efficacy (β = 0.640, p < 0.001). | Financial self-efficacy had a positive effect on perceived detection effectiveness (β = 0.491, p < 0.001). | The variance accounted for (VAF) value of 51.46% indicates complementary partial mediation. | The authors recommend AI monitoring paired with structured training, human-in-the-loop governance, and Shariah-sensitive procedures.
rundown: The analysis used reliability and validity tests, correlation analysis, standardized multiple regression, and regression-based mediation analysis. The direct effect of AI adoption on perceived detection effectiveness stayed significant after accounting for self-efficacy (β = 0.296, p = 0.001).

The indirect effect through financial self-efficacy was 0.314 (Sobel z = 4.806, p < 0.001). Because both the direct and indirect effects were significant, the authors describe the mediation as complementary and partial.

The authors frame the work as addressing a behavioral and institutional gap in the literature: how AI-supported monitoring becomes effective regulatory action through employees' confidence in interpreting and using AI-generated outputs.
sources:
- peer_reviewed | Discover Sustainability | https://doi.org/10.1007/s43621-026-04775-5 | 2026-10-20
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

274749f7459eaf2df169dafc4fcc848f64694721af5522b26e41abc260c4362c

previous

0000000000000000000000000000000000000000000000000000000000000000