Algorithmic Governance and Audit Efficiency: The Role of AI Adoption and IT Governance in Jordanian Commercial Banks

Algorithmic Governance and Audit Efficiency: The Role of AI Adoption and IT Governance in Jordanian Commercial Banks | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Algorithmic Governance and Audit Efficiency: The Role of AI Adoption and IT Governance in Jordanian Commercial Banks Iman Babiker, Fawwaz Alrwabdah, AHMAD Alomari, Mohammed Ahmed Mohammed Ali Ramadan, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9348015/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract AI is rapidly developing in auditing practices worldwide, but there is little empirical evidence on the link between AI and audit efficiency among emerging market banks. In addition, the organizational variables through which AI improves audit efficiency— for example, IT governance maturity— are not theoretically and empirically theoretically sound and empirically untested. This study explores the effect of AI adoption on audit delay in Jordanian banks and the altering role of IT governance maturity. The data for 2015–2024 are summarized in a balanced panel of 15 commercial and Islamic banks listed on the ASE, indicating 150 bank years. The normalized score of AI adoption is obtained through bilingual content analysis of annual reports, and a CoBIT-aligned composite index for IT governance maturity is measured by a standardized score of the bilingual content analysis of the annual reports. Four model specifications employ pooled OLS regression with Driscoll–Kraay standard errors robust to heteroskedasticity, autocorrelation, and cross-sectional dependence. This suggests that AI adoption significantly reduces audit delay (= 26.07, p = 0.01); an increased AI score = 4.06 fewer audit days, which translates into a 6.5% efficiency improvement; and an increase in one standard deviation in the AI score yields 4.06 fewer audit days, or a 6.5% improvement in accuracy. I conclude that the interaction between AI adoption and IT governance maturity is negative and significant: 3 = 6.22, p = 0.011, and banks with larger IT governance maturity significantly increase the efficiency of audit performance. The results are reliable when the year fixed effects are replaced with a COVID-19 control variable. The present study contributes the first archival evidence from emerging-market banks in the MENA region on the AI–audit efficiency nexus, incorporates Agency Theory, TOE and Institutional Theory, and emphasizes IT governance as a critical barrier to achieving an efficient use of AI in auditing. JEL Classification: M41, M42, G21, G34, O33 Business and commerce/Business and management Social science/Business and management Business and commerce/Information systems and information technology Physical sciences/Mathematics and computing Audit Efficiency AI Adoption IT Governance Driscoll-Kraay Jordanian Banks Panel Data Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction The global auditing profession is undergoing a fundamental transformation fueled by AI, robotic process automation, and advanced data analytics. In addition to conducting full-scale testing of full population, real-time anomaly detection and automated reconciliation, major international audit firms have invested billions of dollars in AI-powered platforms for full population testing, real-time anomaly detection and automated reconciliation (Lin & Maginnis 2025 ; Eisikovits, Johnson & Markelevich, 2025 ). These technologies suggest changing the efficiency equation of audit engagement by compressing the time needed for normal procedures and allowing auditors to redirect cognitive efforts to higher-order judgment tasks. Yet the empirical evidence on whether AI adoption improves measurable audit outcomes in particular in the emerging market, in areas such as emerging markets where institutional and governance infrastructure differ substantially from those of developed economies (Agostino, Lourenço, Jorge, Bracci & Cruz, 2025 ). The question of audit efficiency is far from being academic. Account delay, defined as the distance from the fiscal year-end to the date an independent auditor is released by the accounting firm, is a direct and visible signal of auditor effectiveness in the audit process (Rahman, Zhu & Yue, 2024 ; Tan, Chang, Zheng & Chan, 2025 ). For banks under regulatory regimes that require a hard filing deadline, changes in delay do not have significant economic and signaling impacts, as early filing banks signal operational performance and confidence with investors, regulators, and depositors, and may also signal that delays might be causing financial complexity or increased audit risk. In Jordan, all banks must submit their audited financial statements by March 31 in the fiscal year-end, a window in which audit delays are particularly interesting. Jordan is an interesting environment to study the impact of AI on audit efficiency. Jordanian banks include 15 commercial and Islamic banks listed on the Amman Stock Exchange (ASE), supervised by the Central Bank of Jordan (CBJ), and subject to Basel III requirements. In the National Financial Inclusion Strategy issued in 2018, the CBJ encouraged technological inclusion through FinTech and digital innovation. This encouragement resulted in the adoption of AI tools in the financial sector, with larger banks, such as the Arab Bank and the Housing Bank for Trade and Finance, adopting AI first in 2019–2021, followed by smaller Islamic banks in 2022–2023 (Neiroukh & Caglar, 2025 ). The staggered adoption of AI technology, partially caused by encouragement from regulators and partially caused by the digitization shock of the COVID-19 pandemic, provides variation for identification. Although the preconditions described above seem to be met in this field, research on AI use in the audit process is still at a very early stage. Several studies have provided a framework for AI applications in different audit processes and suggested empirical investigation (e.g., Munoko, Brown-Liburd & Vasarhelyi, 2020 ; Lehner, Ittonen, Silvola, Ström & Wuhrleitner, 2022) or discussed how individuals perceive AI use in auditing (e.g., Gambhir, Srivastava & Gupta, 2025 ). However, few studies have provided empirical evidence. For example, Tan et al. ( 2025 ) document from China that auditors’ clients’ AI use increases audit quality and reduces the audit lag; Rahman, Islam, Gupta & Alabsy (2024) show that auditors’ as well as their clients’ use of AI decreases audit report lag; and Lai ( 2025 ) documents that auditors’ adoption of AI technology reduces the audit fees through the reduction of information asymmetry in the audit process. However, to the best of our knowledge, archival evidence on AI use in the audit process in the MENA region is lacking, and the role of IT governance remains unexplored. This is an important gap, as the benefits of AI implementation are not unconditional. The technological, organizational, and environmental (TOE) context of an organization, such as IT governance and data infrastructure, may play an influential role in reaping the benefits of the capabilities of new technologies to improve performance (Marhraoui, 2026 ). IT governance may be an important facilitator of auditing efficiency. Almaqtari ( 2024 ) documents that IT governance is important to address risks and complexities in using AI in accounting and auditing, and Hu, Chen, Hsu, and Tzeng ( 2023 ) identify AI governance and data infrastructure as vital aspects of a successful AI-based audit framework. The research questions of this study are as follows: (1) Does AI adoption promote audit efficiency in Jordanian commercial banks? (2) Is this impact moderated by the maturity of IT governance? To analyze this research question, we used the Pooled OLS estimation technique with Driscoll-Kraay standard errors, which is robust to heteroskedasticity, autocorrelation, and cross-sectional correlation (Driscoll & Kraay, 1998 ; Hoechle, 2007 ) using Python (Biesialski et al., 2009) through (PyCharm, 2023). This study offers several contributions. This is the first study to provide archival evidence of AI adoption and overall audit efficiency for a MENA region bank (an emerging market). The study contributes IT governance as a moderating variable with a strong theoretical background. It contributes an integrated Agency Theory, TOE, and Institutional Theory framework. This study demonstrates the methodology for conducting bilingual content analysis for AI adoption proxies in the MENA region. The remainder of this study is organized as follows: Section 2 discusses the theoretical framework and hypotheses, and Section 3 explains the methodology. Section 4 presents the results, and Section 5 discusses the findings. Finally, Section 6 concludes the paper. 2. Theoretical Framework and Hypothesis 2.1 Theoretical Foundations 2.1.1 Agency Theory and Algorithmic Monitoring Ultimately, the reason for audit efficiency concerns from the standpoint of Agency Theory and how AI technology could potentially help with it becomes clear. The disconnection between ownership and control results in information asymmetry in the relationship between principals and agents (Jensen & Meckling, 1976 ; Fama & Jensen, 1983 ). External audits can act as a monitoring process to reduce the severity of this asymmetry. The possibility of full-population testing and temporal compression via near-real-time analysis are two mechanisms through which AI tools redefine audit IT monitoring (Munoko et al., 2020 ; Li et al., 2024 ; Lehner et al., 2022 ). Tied to agency theory, the adoption of AI is expected to increase audit efficiency by decreasing the time and effort required to reach a certain level of assurance. 2.1.2 Technology-Organization-Environment (TOE) Framework According to the TOE framework, the successful adoption of technology is a function of technological readiness, organizational or operational capacity, and environmental influences such as need or pressure (Tornatzky & Fleischer, 1990 ; Marhraoui, 2026 ). In the organization dimension, outputs may be based on IT governance, which functions to “decide which AI projects get prioritized, resourced and evaluated among competing ones” (Almaqtari, 2024 ; Emett, Eulerich, Pikoos, & Wood, 2026 ). Schiff, Kelley, and Ibáñez ( 2024 ) determined that the use of AI by IT auditors has some technical limitations that can be managed through IT governance. It is based on the TOE in which IT governance is hypothesized as a moderator. 2.1.3 Institutional Theory Institutional Theory explains the exogenous factors encouraging AI adoption, including coercive elements relating to CBJ regulations, mimetic factors in terms of competitor imitation, and normative factors linked to professional standards (DiMaggio & Powell, 1983 ; Scott, 2014 ; Bonsu, Wang & Guo, 2023 ). Roncato, Medeiros, and Lerner ( 2026 ) found that the influence of regulatory obligations positively impacts the adoption and compliance quality of governance practices in emerging markets. 2.2 Literature Review 2.2.1 AI Adoption and Audit Efficiency While the overall literature on the role of AI in auditing is still limited, as previously mentioned, some scholars have moved beyond conceptual frameworks to empirical analyses. For instance, Munoko, Buck, and Tumunya (2020) provided a seminal analysis of the opportunities and risks of AI in auditing. Based on 25,408 firm-year samples in China, Tan, Cheung, and Yu (2025) empirically measured how using AI in auditing improves audit quality and reduces audit lags. Rahman, Chowdhury, and Islam (2024) reported from their empirical study that when AI is used jointly, audit report lag is significantly reduced. In a similar context, Lai ( 2025 ) empirically documented that AI reduces audit fees by reducing information asymmetry. Fawzi Khan, Hiba Jan, and Samina Zia-ul-haq (2025) found that AI adoption empowers integrated financial reporting in the Gulf Cooperation Council markets, and this relationship is mediated by audit quality. Through a controlled experiment conducted in Egypt, Abouelela et al. ( 2025 ) confirmed the role of AI in enhancing audit planning. Peng, Gong, and Zhong ( 2026 ) reported how AI improves the quality of accounting information. Fajardo and Neiva ( 2025 ) found in their empirical analysis that machine learning algorithms significantly improved auditing in the Brazilian government sector. Finally, Agostino et al. ( 2025 ) confirm the vast shortage of empirical studies on algorithmic technologies in accounting. 2.2.2 IT Governance as a Moderating Mechanism Almaqtari ( 2024 ), surveying 228 Saudi IT professionals, identified IT governance as a significant predictor of the extent to which companies integrate AI into their operations. Hu, Xie, Cheng, Liu, and Zhang (2023) contend that AI governance is a key factor in the success of AI-powered auditing practices. Emett, Feldman, and Paim (2026) present a governance framework validated through interviews and focus groups that include 69 control considerations. Lacmanović and Skare ( 2025 ) analyzed several companies and showed that auditing AI for bias is not a standardized process. Silic, Silic, and Kind-Trüller ( 2025 ) conceive shadow AI as an instance of governance failure. Neiroukh and Caglar ( 2025 ) prove, through interviews with managers in Jordanian banks, that AI and IT governance positively impact the quality of information systems. Li and Goel ( 2025 ) identify data governance and model explainability as critical requirements for AI auditability. von Zahn, Zacharias, Lowin, Chen, and Hinz (2025) provide a design framework for conformity assessments involving AI with components of fairness and explainability. 2.3 Hypothesis Development H1: AI adoption is negatively associated with audit delay in Jordanian commercial banks, such that banks with higher AI adoption scores exhibit shorter audit delays, ceteris paribus . The direct relationship is built from the monitoring function of Agency Theory and supported by Tan et al. ( 2025 , Rahman et al. ( 2024 , Lai 2025 , Peng et al. ( 2026 ). If these AI tools enable data extraction, reconciliation, and confirmation matching, they are expected to reduce the audit process time. H2: IT governance maturity positively moderates the relationship between AI adoption and audit efficiency, such that the negative association between the AI adoption score and audit delay is stronger for banks with higher IT governance maturity . The moderating effect stems from the TOE and has been supported by Almaqtari ( 2024 ), Neiroukh and Caglar ( 2025 ), and Hu et al. ( 2023 ). Mature IT governance is correlated with a robust data infrastructure foundation and accountability that allow investments in AI to have a higher probability of leading to efficiency improvements. 3. Research Methodology 3.1 Research Design and Sample This study is guided by a positivist philosophy and should be understood as deductive in nature (Bryman, 2016 ). The sample covers the entire population of all commercial and Islamic banks present in the ASE during the period from to 2015–2024, which totals 15 banks each year, contributing to a balanced panel of 150 bank-year observations. The data comprise audited financial statements, corporate governance disclosures, and hand-collected audit reports. 3.2 Variable Measurement Dependent Variable: Audit Delay. It is the number of calendar days between the fiscal year-end (December 31) and the audit report, an audit timeliness proxy that is often related to audit efficiency (Knechel & Sharma, 2012 ; Tan et al., 2025 ; Rahman et al., 2024 ; Abouelela et al., 2025 ). The independent Variable: AI Adoption Score. Developed by the researcher based on bilingual content analysis guided by Loughran and McDonald ( 2011 ). Machine learning, artificial intelligence, process automation, and data analytics. It was normalized for report length and scaled from [0,1]. This measure has been tested for measuring the intensity of technology adoption (Peng, Tong, and Xie, 2026; Xie, Peng, & Tong,2026; Li and Chen, 2025 ). Moderating Variable: IT Governance. A COBIT-based composite index (0–1) from strategic IT alignment, IT risk management, IT resource management, IT performance measurement, IT compliance, IT value delivery, and data governance (Almaqtari, 2024 ; Hu et al., 2023 ; ISACA, 2019 ). Control Variables. Revenue growth represents the operational complexity (Knechel and Sharma 2012 ; Min et al. 2025 ). Big4 accounts for audit firm resources and technology ( Abouelela et al., 2025 ; Saif-Alyousfi, 2025 ). ROA was used by Daugherty and Dickins ( 2010 ) and Xie et al. ( 2026 ) to proxy profitability and risk. Use represents the financial risk (Khan et al., 2025 ; Nguyen, 2025 ). Continuous variables are winsorized at the 1st/99th percentiles (Kothari, Leone, and Wasley, 2005 ). Table 1 Variable Definitions and Measurement Variable Measurement Role Source AI_Score Normalized AI keyword frequency (0–1) Independent Annual reports Audit_Delay Days: fiscal year-end to audit report Dependent Audit reports IT_Gov COBIT-aligned index (0–1) Moderator Governance filings Growth Year-on-year revenue growth Control Financial statements Big4 1 if Big Four auditor Control Audit reports ROA Net income / total assets Control Financial statements Leverage Total liabilities / total assets Control Financial statements 3.3 Econometric Specification Model 1 (Baseline): \(\:Audit\_Delay\:=\:\beta\:₀\:+\:\beta\:₁AI\_Score\:+\:\beta\:₂Growth\:+\:\beta\:₃Big4\:+\:\beta\:₄ROA\:+\:\beta\:₅Leverage\:+\:\delta\:ₜ\:+\:\epsilon\:\) Model 2 (Direct): \(\:Audit\_Delay\:=\:\beta\:₀\:+\:\beta\:₁AI\_Score\:+\:\beta\:₂IT\_Gov\:+\:\gamma\:{\prime\:}Controls\:+\:\delta\:ₜ\:+\:\epsilon\:\) Model 3 (Interaction): \(\:Audit\_Delay\:=\:\beta\:₀\:+\:\beta\:₁AI\_Score\:+\:\beta\:₂IT\_Gov\:+\:\beta\:₃(AI\_Score\times\:IT\_Gov)\:+\:\gamma\:{\prime\:}Controls\:+\:\delta\:ₜ\:+\:\epsilon\:\) Model 4 (Robustness): specification with COVID dummy replacing year-fixed effects. \(\:Audit\_Delay\:=\:\beta\:₀\:+\:\beta\:₁AI\_Score\:+\:\beta\:₂IT\_Gov\:+\:\beta\:₃(AI\_Score\times\:IT\_Gov)\:+\:\gamma\:{\prime\:}Controls\:+\:COVIDₜ\:+\:\epsilon\:\) Time-fixed effects year (δₜ) absorb time-varying shocks. The interaction term was mean-centered to reduce multicollinearity (Aiken & West, 1991 ). In case of detected cross-sectional dependence (Pesaran CD test), all models are estimated with Driscoll-Kraay standard errors, robust to heteroskedasticity, autocorrelation, and cross-sectional dependence (Driscoll & Kraay, 1998 ; Hoechle, 2007 ; Vogelsang, 2012 ). However, with N = 15 banks, it is not safe to use cluster-robust SE (Cameron et al., 2008 ). All analyses were performed using the PyCharm IDE and Python 3.12. 4. Empirical Results 4.1 Descriptive Statistics The descriptive statistics are presented in Table 2 . The average audit delay is 62.39 days (s.d. = 6.86) and ranges between 48 and 75 days, in accordance with the CBJ’s deadline of March 31. The mean adoption score is 0.178, with a standard deviation of 0.155, with banks reporting scores from virtually no AI adoption to 0.576 adoption, indicating moderate levels of adoption at the time of data collection but a general upward trend. The IT governance composite had a mean of 0.500 (SD = 0.250) and ranged from 0.000 to 1.000, illustrating a high degree of variability. For controls, the percentage of bank-years that were Big Four audited was 71.3%, the average ROA was 5.02%, and average use was 34.41%. Table 2 Descriptive Statistics Variable N Mean Median SD Min P25 P75 Max Audit Delay (days) 150 62.3885 63 6.8550 48 58 67 75 AI Adoption Score 150 0.1783 0.1271 0.1549 0.0068 0.0474 0.2776 0.5762 IT Governance 150 0.5000 0.4810 0.2500 0.0000 0.3257 0.6532 1.0000 Revenue Growth 150 0.0920 0.0891 0.1192 -0.1703 0.0069 0.1816 0.3488 Big 4 Auditor 150 0.7133 1.0000 0.4537 0.0000 0.0000 1.0000 1.0000 Return on Assets 150 0.0502 0.0486 0.0396 -0.0396 0.0220 0.0791 0.1377 Financial Leverage 150 0.3441 0.3360 0.1538 0.0500 0.2268 0.4647 0.6952 4.2 Correlation Analysis and Multicollinearity The Pearson correlation coefficient with the variance inflation factor (VIF) is presented in Table 3 . AI adoption and audit delay have a significantly high negative correlation (− 0.765), supporting H1. IT governance is correlated with audit delay at a value of − 0.485. The highest correlation among the independent variables was between AI_Score and IT_Gov, with a coefficient of 0.303, which is below the accepted level of 0.70 (Kennedy, 2008 ). The VIF values lie between 1.02 and 1.12, indicating that there is no multicollinearity. Table 3 Pearson Correlation Matrix with VIF Variables AI_Score IT_Gov Growth Big4 ROA Leverage VIF AI_Score 1.000 0.303*** 0.074 -0.117 0.089 -0.001 1.12 IT_Gov 0.303*** 1.000 0.048 -0.096 0.078 -0.019 1.11 Growth 0.074 0.048 1.000 0.128 0.123 0.026 1.04 Big4 -0.117 -0.096 0.128 1.000 0.099 0.080 1.05 ROA 0.089 0.078 0.123 0.099 1.000 0.108 1.05 Leverage -0.001 -0.019 0.026 0.080 0.108 1.000 1.02 4.3 Diagnostic Tests The panel diagnostic tests are presented in Table 4 . The Breusch-Pagan (LM = 13.73, p = 0.546) and White tests (LM = 73.82, p = 0.877) do not reject the homoskedasticity. The Durbin-Watson statistic is 2.29, and the Wooldridge AR(1) test results F = 2.83 (p = 0.095), indicating mild autocorrelation. However, the Pesaran CD test rejects cross-sectional independence (CD = − 2.19, p = 0.029), which is why Driscoll-Kraay standard errors are used. Table 4 Panel Diagnostic Tests Test Statistic p-value Decision Breusch-Pagan 13.733 0.546 Fail to reject H₀ White 73.818 0.877 Fail to reject H₀ Wooldridge AR(1) 2.832 0.095 Fail to reject H₀ Pesaran CD −2.188 0.029 Reject H₀ ** 4.4 Regression Results Table 5 reports the results of all four model specifications with Driscoll-Kraay standard errors. Table 5 Regression Results (Driscoll-Kraay Standard Errors) Variable Model 1 Model 2 Model 3 Model 4 (Robust.) AI_Score −26.220*** −26.780*** −26.071*** −24.275*** (1.801) (2.120) (2.006) (2.200) IT_Gov −7.530*** −7.516*** −7.453*** (0.651) (0.599) (0.573) AI_Score × IT_Gov −6.221** −9.804*** (2.397) (2.064) Growth −11.109*** −10.568*** −10.222*** −9.688*** (2.289) (1.163) (1.308) (1.132) Big4 −3.143*** −3.604*** −3.591*** −3.382*** (0.435) (0.460) (0.451) (0.435) ROA −23.134*** −21.569*** −21.365*** −21.948*** (3.353) (5.012) (4.815) (4.391) Leverage 9.578*** 9.629*** 9.723*** 9.420*** (2.174) (1.692) (1.658) (1.675) COVID −2.163*** (0.543) Year FE Yes Yes Yes No N 150 150 150 150 R² 0.763 0.826 0.827 0.815 Adj. R² 0.738 0.807 0.807 0.805 Notes : *** p < 0.01 , ** p < 0.05 , * p < 0.10 . The standard errors are in parentheses. Models1-3 include year effects, and Model 4 uses a COVID dummy (1 for 2020–2024). Model 1 shows that AI adoption has a significant negative relationship with audit delay ( = -26.22, p = 0.01). The increase in the AI score (0.155) of one standard deviation translates to approximately 4.06 less or a 6.5% better score compared to 62.39 days, which is the average. All controls are significant: Growth (β = −11.11, p < 0.01), Big4 (β = −3.14, p < 0.01), ROA (β = −23.13, p < 0.01), and Leverage (β = 9.58, p < 0.01). R² = 0.763. Model 2 adds IT governance, substantially improving the fit(R 2 = 0.826). IT governance is inversely significant ( = -7.53, p < 0.01), and AI adoption is not significantly important ( = -26.78, p < 0.01), proving the independent impacts on audit efficiency. Model 3 adds the interaction term, which is negative and substantial ( 3 = -6.22, p = 0.011), which directly validates H2. It can be seen in a simple slopes analysis that the marginal AI effect for less than 24.52 at the low IT governance (− 1 SD) and above 27.63 at the high IT governance (+ 1 SD) is strong. Model 4 (robustness) the year FE is replaced by a COVID dummy. With a very large magnitude, the interaction is very relevant (β 3 = -9.80, p < 0.01). COVID was noteworthy and negative (216 = -, p < 0.01), which proves the acceleration of digitization during the pandemic. 4.5 Hypothesis Testing Summary H1 (Supported) AI adoption is significantly and negatively associated with audit delay across all four models (p < 0.01), with coefficients ranging from − 24.28 to − 26.78. The adoption trajectory from near-zero (2015) to 0.41 (2024) implies approximately 10.7 fewer days, a 17% efficiency gain. H2 (Supported) The interaction term is significant in Model 3 (β₃ = −6.22, p = 0.011) and Model 4 (β₃ = −9.80, p < 0.01). Banks with high IT governance extract significantly greater efficiency gains from their AI investments. 5. Discussion 5.1 AI Adoption and Audit Efficiency The finding that AI adoption significantly reduces audit delay complements and contributes to the sparse existing literature. Our findings (β = −26) are comparable and consistent with those of Tan et al. ( 2025 ), who examined the effect of AI adoption on audit lag in China and concluded that AI reduces audit lag. Our study is based on the context of Jordan, where AI adoption is still in its infancy; thus, an increase in adoption yields a marginal but significant improvement in the efficiency of accounting firms’ operations as they shift from manual to automated processes. Our findings further support Bonsu et al. ( 2023 ), who reported that fintech adoption enhances accounting practices in emerging markets, and Peng et al. ( 2026 ), who found that AI improves accounting information. This is particularly relevant given the audit technology environment in Jordanian banks. While audits in developed countries are established within a technological context, banks in emerging markets such as Jordan are in the process of shifting from a manual environment to a technological one, thereby creating more significant efficiency implications. Fajardo and Neiva ( 2025 ) examine the efficiency effects of artificial intelligence in the audit records of the public sector in Brazil and find similar results. 5.2 The Moderating Role of IT Governance The significance of the interaction term indicates that IT governance enhances the efficiency of AI. The significance of this interaction term supports the organizational aspect of the TOE framework: organizations must do something with their technological capabilities to derive value. This finding supports Almaqtari’s ( 2024 ) survey-based findings on IT value creation but advances this and other extant AI literature by using archival rather than perceptual data and by focusing on a specific value outcome instead of a generic measure. This finding is also consistent with Neiroukh and Caglar’s ( 2025 ) study of Jordanian banks, which found that AI and governance positively affect the quality of information. The moderation plot shows that when IT governance is low, the slope for AI and audit delay is flatter, but it steeply increases when IT governance is high (Fig. 4 ). This result supports Hu et al.’s ( 2023 ) finding that improving AI governance is a top priority and Emett et al.’s ( 2026 ) argument that governance structures are needed to mitigate the risks of deploying AI. Both Lacmanović and Skare ( 2025 ) and von Zahn et al. ( 2025 ) reinforce this need for standardized governance frameworks. 5.3 Control Variable Findings All the control variables are statistically significant and expected. The negative relationship with growth may capture the operational dynamism that accompanies the growth phase (Min et al., 2025 ). The impact of the Big Four on audit report lag (approximately 3.5 fewer days) is in line with superior technological infrastructure and supports Abouelela et al.’s ( 2025 ) findings. Return on assets (ROA) negatively affects audit report lag, confirming that audits of profitable banks are less risky and require less effort (Xie et al., 2026 ). In contrast, use has a positive effect, indicating that audits of highly leveraged banks are a focus (Khan et al., 2025 ). The dummy variable for the COVID pandemic is statistically significant, and this result verifies that the pandemic triggered digitization efforts (Albous et al., 2025 ). 6. Conclusion 6.1 Summary This study empirically tests the impact of artificial intelligence (AI) adoption on audit efficiency and the moderating role of information technology (IT) governance in Jordanian banks. The analysis was based on 150 bank-year observations from all 15 banks listed on the Amman Stock Exchange (ASE) from 2015 to 2024. The results show that the adoption of AI significantly reduces audit delay; IT governance reduces audit delay and enhances AI’s efficiency effects; and the results are controlled for the COVID-19 period and several model modifications. 6.2 Theoretical Contributions Theoretically, this study contributes to the literature in three dimensions. First, this study adds to Agency Theory by extending it to algorithmic monitoring. The findings reveal that AI serves as an information source to assist in the governance process by collapsing the monitoring time element instead of providing situational information, as established in previous studies (Lehner et al., 2022 ; Gorwa, Binns & Katzenbach, 2020 ). Second, this study adds to the TOE framework by conceptualizing IT governance as a moderating organizational capability, not a direct antecedent, in the chain linking IT resources to organizational performance. Third, this study integrates Institutional Theory to explain the role of coercive, mimetic, and normative motives behind MENA banks’ adoption and governance practices of AI technology (Roncato et al., 2026 ; Albous et al., 2025 ). 6.3 Practical Implications For bank managers, AI investment will not bring efficiency benefits unless accompanied by a suitable investment in IT governance. For the CBJ, it may be wise to continue fostering AI adoption and governance maturity. For audit firms, clients’ AI maturity and governance strength should be included as part of the overall engagement strategy. Finally, for MENA policymakers, structured regulatory encouragement can lead to positive AI adoption in the region, provided that governance issues are addressed. 6.4 Limitations and Future Research The study’s limitations are its relatively small sample size (15 banks), AI measurement based on content analysis, binary scoring of IT governance, and sole use of an efficiency proxy. Future research should consider a larger sample from the MENA region, use patent-based AI measures alongside textual-based measures (Ding & Gu, 2026 ), employ a more nuanced IT governance measure, and consider additional outcome variables such as audit fees and the probability of financial statement restatements. To firmly establish causality, an instrumental variable methodology based on specific regulatory events should be used. Declarations Data Availability The data used to support these conclusions are derived from the annual reports and financial statements of companies traded on the Amman Stock Exchange (ASE). The results and analyses of the datasets presented in the present study can be accessed from the authors of this study upon reasonable request. Acknowledgements Princess Nourah bint Abdulrahman University Researchers Supporting Project number ( PNURSP2026R861 ), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia. Competing Interests The authors agree that this research was conducted in the absence of any self-benefit or commercial or financial conflicts. Ethical Statements This study uses publicly available secondary data from annual and financial reports. No human participants were involved; therefore, ethical approval was not required. References Abouelela O, Diab A, Saleh S (2025) The relationship between AI and audit planning in emerging economies. SAGE Open, 15(4) Acemoglu D, Restrepo P (2019) Automation and new tasks are also important. J Economic Perspect 33(2):3–30 Agostino D, Lourenço R, Jorge S, Bracci E, Cruz I (2025) Data science and public sector accounting. Public Money & Management Aiken LS, West SG (1991) Multiple regression: Testing and interpreting interactions. Sage Albous MR, Stephens M, Al-Jayyousi OR (2025) AI and the GCC workforce. Humanit Social Sci Commun 12(1):1649 Almaqtari FA (2024) IT governance in the integration of AI in accounting and auditing. Economies 12(8):199 Bao HX, Liu WF, Dai Z (2025) AI vs. public administrators. Technol Forecast Soc Chang 215:124102 Bonsu MOA, Wang Y, Guo YS (2023) Does Fintech lead to better accounting practices? Acc Res J 36(2/3):129–147 Bryman A (2016) Social Research Methods, 5th edn. Oxford University Press Cameron AC, Gelbach JB, Miller DL (2008) Bootstrap-based improvements for inference with clustered errors. Rev Econ Stat 90(3):414–427 Chai L et al (2025) AI-generated content and ESG. Romanian J Economic Forecast 28(3):5–23 Crawford J, Nilsson F (2023) Integrating ESG risks into control and reporting. Handbook of Big Data and Analytics in Accounting and Auditing. Springer, pp 255–277 Daugherty B, Dickins D (2010) Perceptions of auditor independence and financial reporting quality. Adv Acc Behav Res 13:169–194 DiMaggio PJ, Powell WW (1983) The iron cage revisited. Am Sociol Rev 48(2):147–160 Ding H, Gu HY (2026) Supply chain shared auditor and corporate AI adoption. Asia Pacific Journal of Marketing and Logistics Driscoll JC, Kraay AC (1998) Consistent covariance matrix estimation with spatially dependent panel data. Rev Econ Stat 80(4):549–560 Eisikovits N, Johnson WC, Markelevich A (2025) Should accountants be afraid of AI? Acc Horizons 39(2):117–123 Emett S, Eulerich M, Pikoos J, Wood DA (2026) Generative AI governance framework. Accounting Horizons Fajardo B, Neiva S (2025) AI application for auditing in the Brazilian federal government. Public Money & Management Fama EF, Jensen MC (1983) Separation of ownership and control. J Law Econ 26(2):301–325 Gambhir B, Srivastava A, Gupta N (2025) AI in audit and accounting: A qualitative exploration. Asian J Acc Gov, 24 Gorwa R, Binns R, Katzenbach C (2020) Algorithmic content moderation. Big Data Soc, 7(1) Harymawan I et al (2023) CEO facial masculinity and tax avoidance. Cogent Bus Manage 10(1):2171644 Henderson MD (2025) Agentic AI and the ethics of leadership maintenance. Leadership & Organization Development Journal Hoechle D (2007) Robust standard errors for panel regressions with cross-sectional dependence. Stata J 7(3):281–312 Hu KH, Chen FH, Hsu MF, Tzeng GH (2023) Governance of AI in a business audit. Financial Innov 9(1):117 ISACA (2019) COBIT 2019 Framework. ISACA Islam MS et al (2026) AI-enabled environmental auditing and sustainability disclosure. Corporate Social Responsibility and Environmental Management Jensen MC, Meckling WH (1976) Theory of the firm. J Financ Econ 3(4):305–360 Jlifi B et al (2026) Reconciling divergence among ESG scores. Computational Economics Kennedy P (2008) A guide to econometrics, 6th edn. Blackwell Khan F, Jan SU, Zia-ul-haq HM (2025) AI adoption, audit quality and integrated financial reporting in GCC. Asian Rev Acc 33(3):464–495 Knechel WR, Sharma DS (2012) Auditor-provided nonaudit services and audit effectiveness. Auditing: J Pract Theory 31(4):73–105 Kothari SP, Leone AJ, Wasley CE (2005) Performance matched discretionary accrual measures. J Account Econ 39(1):163–197 Lacmanović S, Skare M (2025) AI bias auditing. Rev Acc Finance 24(3):375–400 Lai J (2025) AI applications and audit fees. Int Rev Econ Finance 103:104421 Lehner OM et al (2022) AI-based decision-making in accounting and auditing. Acc Auditing Account J 35(9):109–135 Li GH, Chen YT (2025) Digitally powered ESG evaluation. Romanian J Economic Forecast 28(3):89–109 Li N, Kim M, Dai J, Vasarhelyi MA (2024) Using AI in ESG assurance. J Emerg Technol Acc 21(2):83–99 Li YQ, Goel S (2025) Bridging IT auditors and AI auditing. Adv Acc 69:100842 Lin TLJ, Maginnis J (2025) Generative AI and edge AI in auditing. Current Issues in Auditing Loughran T, McDonald B (2011) When is a liability not a liability? J Finance 66(1):35–65 Marhraoui MA (2026) Drivers and outcomes of AI adoption. IEEE Trans Eng Manage 73:920–935 Min H, Mirza SS, Huang CM (2025) Digital transformation and audit opinions. Journal of Corporate Accounting and Finance Munoko I, Brown-Liburd HL, Vasarhelyi M (2020) Ethical implications of AI in auditing. J Bus Ethics 167(2):209–234 Neiroukh N, Caglar D (2025) Information systems quality and corporate sustainability. Systems 13(7):537 Nguyen TA (2025) Bibliometric analysis of ESG risk. Bus Strategy Dev, 8(3), e70163 Peng Z, Gong YT, Zhong X (2026) Can AI improve accounting information quality? Applied Economics Rahman MJ, Ziru A (2023) Clients' digitalization and audit quality. Int J Acc Inform Manage 31(2):221–246 Rahman MJ, Zhu HT, Yue L (2024) AI adoption, audit quality and efficiency. Managerial Auditing J 39(6):668–699 Roncato S, Medeiros LM, Lerner AF (2026) Regulation and corporate governance practices. J Financial Regul Compliance 34(2):177–197 Saif-Alyousfi AYH (2025) AI, information environment, and capital market efficiency. Res Int Bus Finance 79:103094 Schiff DS, Kelley S, Ibáñez JC (2024) AI ethics auditing. Big Data Soc, 11(4) Scott WR (2014) Institutions and organizations, 4th edn. Sage Silic M, Silic D, Kind-Trüller K (2025) From shadow IT to shadow AI. Strategic Change Tan JH, Chang SM, Zheng Y, Chan KC (2025) Client AI and audit quality. Int Rev Financial Anal 104:104271 Tornatzky LG, Fleischer M (1990) The processes of technological innovation. Lexington Books Vogelsang TJ (2012) Heteroskedasticity, autocorrelation, and spatial correlation robust inference. J Econ 166(2):303–319 von Zahn M et al (2025) Navigating AI conformity. Electron Markets 35(1):24 Xie LF, Peng ZB, Tong XG (2026) AI strategy, earnings management, and corporate fraud. Econ Model 156:107460 Zhao X, Tong YS, Lee H, Shahzad U (2025) AI in enhancing corporate environmental disclosure. Energy Econ 148:108680 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 16 Apr, 2026 Editor assigned by journal 15 Apr, 2026 Submission checks completed at journal 14 Apr, 2026 First submitted to journal 07 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9348015","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":619121885,"identity":"97cf70e4-f936-4459-84cc-cbf12433ed47","order_by":0,"name":"Iman Babiker","email":"","orcid":"","institution":"Princess Nourah bint Abdulrahman University","correspondingAuthor":false,"prefix":"","firstName":"Iman","middleName":"","lastName":"Babiker","suffix":""},{"id":619121886,"identity":"2bec9d8c-530e-4bf9-bc90-d15b86a75dfd","order_by":1,"name":"Fawwaz Alrwabdah","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYBAC9gYYS4LBgJmhwkaOgYEHvxaeAyhazqQZk6iFse1wYgNBLexnDD/+3GOX2D+7eePjgjNp6RuOnz344AODnZxuAw4tPDnGEhLPkhNn3DlWbDyjwiZ3w5m8ZMMZDMnGZgewa7FnyN0gYXCAObfhRo6ZNM+ZtNwNB0AMhgOJ23Bo4eF/u/lHwoH63Pk3csx/87YdTjc4/4aAFoncbRIHDhzO3QC0hRmoJcHgBiFbJN5/s2w4cLx+I9AvIIcZzrzxxthwhgFuv/DwpyXf/HGg2ljudvPGzzwVNvJ853MMH3yosJPDpQUTKIBVGhCrHATkG0hRPQpGwSgYBSMBAAAMkmYQJx9SQAAAAABJRU5ErkJggg==","orcid":"","institution":"Universiti Sains Malaysia","correspondingAuthor":true,"prefix":"","firstName":"Fawwaz","middleName":"","lastName":"Alrwabdah","suffix":""},{"id":619121887,"identity":"e3f67486-daaf-4645-83ae-0971fc1f506e","order_by":2,"name":"AHMAD Alomari","email":"","orcid":"","institution":"Universiti Sains Malaysia","correspondingAuthor":false,"prefix":"","firstName":"AHMAD","middleName":"","lastName":"Alomari","suffix":""},{"id":619121888,"identity":"51f0c589-1f05-4764-9409-aa42044e4656","order_by":3,"name":"Mohammed Ahmed Mohammed Ali Ramadan","email":"","orcid":"","institution":"University of Tabuk","correspondingAuthor":false,"prefix":"","firstName":"Mohammed","middleName":"Ahmed Mohammed Ali","lastName":"Ramadan","suffix":""},{"id":619121889,"identity":"fddf8f5e-007f-4c09-a962-1c442af0f60e","order_by":4,"name":"Amal Alharthi","email":"","orcid":"","institution":"Princess Nourah bint Abdulrahman University","correspondingAuthor":false,"prefix":"","firstName":"Amal","middleName":"","lastName":"Alharthi","suffix":""},{"id":619121892,"identity":"c8425ad8-e2c7-4036-a90e-c8354fb4b272","order_by":5,"name":"Mashael Bakhit","email":"","orcid":"","institution":"Princess Nourah bint Abdulrahman University","correspondingAuthor":false,"prefix":"","firstName":"Mashael","middleName":"","lastName":"Bakhit","suffix":""}],"badges":[],"createdAt":"2026-04-07 17:08:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9348015/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9348015/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106756373,"identity":"35c57d89-a1d7-4fc1-8e98-7bd6a5ed225b","added_by":"auto","created_at":"2026-04-13 08:04:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":142870,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eEvolution of AI Adoption in Jordanian Banks (2015–2024)\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9348015/v1/357a468ed640a22301ad2b05.png"},{"id":106756374,"identity":"444ac9aa-b1fd-435a-b6da-b43ccd356b76","added_by":"auto","created_at":"2026-04-13 08:04:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":211406,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eFigure 3: Pearson Correlation Heatmap\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9348015/v1/236b7a9aa442f853dad5206b.png"},{"id":106960393,"identity":"75354a9e-bb77-46b8-acb0-c57e0f8c8acf","added_by":"auto","created_at":"2026-04-15 09:20:49","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":322546,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eFigure 2: AI Adoption vs. Audit Delay (colored by IT Governance)\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9348015/v1/fa38df80748c3ef6d67da707.png"},{"id":106959740,"identity":"d3c1aef3-e7ee-4f6a-88e3-1ebabd87e117","added_by":"auto","created_at":"2026-04-15 09:14:17","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":300127,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eFigure 4: Moderating Effect of IT Governance on AI–Audit Delay Relationship\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9348015/v1/38ad73158cbae69288e6a06a.png"},{"id":106756377,"identity":"bbd82596-caaa-4de1-a3da-d0c0e4182af2","added_by":"auto","created_at":"2026-04-13 08:04:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":76978,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eFigure 5: Coefficient Plot – Model 3 (Driscoll-Kraay)\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9348015/v1/3c04706ca363581dc97938da.png"},{"id":106756378,"identity":"e165fce3-00d7-475e-9fdf-4a67a76da044","added_by":"auto","created_at":"2026-04-13 08:04:38","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":205605,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eFigure 6: Audit Delay Patterns\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-9348015/v1/8b34c38d6e3826f4ffb9b364.png"},{"id":108490664,"identity":"99201c24-b933-4cdc-85b1-26ee674b8497","added_by":"auto","created_at":"2026-05-05 09:46:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1572016,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9348015/v1/1aee22da-c8ac-4430-97fd-c511a32f3627.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Algorithmic Governance and Audit Efficiency: The Role of AI Adoption and IT Governance in Jordanian Commercial Banks","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe global auditing profession is undergoing a fundamental transformation fueled by AI, robotic process automation, and advanced data analytics. In addition to conducting full-scale testing of full population, real-time anomaly detection and automated reconciliation, major international audit firms have invested billions of dollars in AI-powered platforms for full population testing, real-time anomaly detection and automated reconciliation (Lin \u0026amp; Maginnis \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Eisikovits, Johnson \u0026amp; Markelevich, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These technologies suggest changing the efficiency equation of audit engagement by compressing the time needed for normal procedures and allowing auditors to redirect cognitive efforts to higher-order judgment tasks. Yet the empirical evidence on whether AI adoption improves measurable audit outcomes in particular in the emerging market, in areas such as emerging markets where institutional and governance infrastructure differ substantially from those of developed economies (Agostino, Louren\u0026ccedil;o, Jorge, Bracci \u0026amp; Cruz, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe question of audit efficiency is far from being academic. Account delay, defined as the distance from the fiscal year-end to the date an independent auditor is released by the accounting firm, is a direct and visible signal of auditor effectiveness in the audit process (Rahman, Zhu \u0026amp; Yue, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Tan, Chang, Zheng \u0026amp; Chan, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). For banks under regulatory regimes that require a hard filing deadline, changes in delay do not have significant economic and signaling impacts, as early filing banks signal operational performance and confidence with investors, regulators, and depositors, and may also signal that delays might be causing financial complexity or increased audit risk. In Jordan, all banks must submit their audited financial statements by March 31 in the fiscal year-end, a window in which audit delays are particularly interesting.\u003c/p\u003e \u003cp\u003eJordan is an interesting environment to study the impact of AI on audit efficiency. Jordanian banks include 15 commercial and Islamic banks listed on the Amman Stock Exchange (ASE), supervised by the Central Bank of Jordan (CBJ), and subject to Basel III requirements. In the National Financial Inclusion Strategy issued in 2018, the CBJ encouraged technological inclusion through FinTech and digital innovation. This encouragement resulted in the adoption of AI tools in the financial sector, with larger banks, such as the Arab Bank and the Housing Bank for Trade and Finance, adopting AI first in 2019\u0026ndash;2021, followed by smaller Islamic banks in 2022\u0026ndash;2023 (Neiroukh \u0026amp; Caglar, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The staggered adoption of AI technology, partially caused by encouragement from regulators and partially caused by the digitization shock of the COVID-19 pandemic, provides variation for identification.\u003c/p\u003e \u003cp\u003eAlthough the preconditions described above seem to be met in this field, research on AI use in the audit process is still at a very early stage. Several studies have provided a framework for AI applications in different audit processes and suggested empirical investigation (e.g., Munoko, Brown-Liburd \u0026amp; Vasarhelyi, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Lehner, Ittonen, Silvola, Str\u0026ouml;m \u0026amp; Wuhrleitner, 2022) or discussed how individuals perceive AI use in auditing (e.g., Gambhir, Srivastava \u0026amp; Gupta, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). However, few studies have provided empirical evidence. For example, Tan et al. (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) document from China that auditors\u0026rsquo; clients\u0026rsquo; AI use increases audit quality and reduces the audit lag; Rahman, Islam, Gupta \u0026amp; Alabsy (2024) show that auditors\u0026rsquo; as well as their clients\u0026rsquo; use of AI decreases audit report lag; and Lai (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) documents that auditors\u0026rsquo; adoption of AI technology reduces the audit fees through the reduction of information asymmetry in the audit process. However, to the best of our knowledge, archival evidence on AI use in the audit process in the MENA region is lacking, and the role of IT governance remains unexplored.\u003c/p\u003e \u003cp\u003eThis is an important gap, as the benefits of AI implementation are not unconditional. The technological, organizational, and environmental (TOE) context of an organization, such as IT governance and data infrastructure, may play an influential role in reaping the benefits of the capabilities of new technologies to improve performance (Marhraoui, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). IT governance may be an important facilitator of auditing efficiency. Almaqtari (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) documents that IT governance is important to address risks and complexities in using AI in accounting and auditing, and Hu, Chen, Hsu, and Tzeng (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) identify AI governance and data infrastructure as vital aspects of a successful AI-based audit framework.\u003c/p\u003e \u003cp\u003eThe research questions of this study are as follows: (1) Does AI adoption promote audit efficiency in Jordanian commercial banks? (2) Is this impact moderated by the maturity of IT governance? To analyze this research question, we used the Pooled OLS estimation technique with Driscoll-Kraay standard errors, which is robust to heteroskedasticity, autocorrelation, and cross-sectional correlation (Driscoll \u0026amp; Kraay, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1998\u003c/span\u003e; Hoechle, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) using Python (Biesialski et al., 2009) through (PyCharm, 2023).\u003c/p\u003e \u003cp\u003eThis study offers several contributions. This is the first study to provide archival evidence of AI adoption and overall audit efficiency for a MENA region bank (an emerging market). The study contributes IT governance as a moderating variable with a strong theoretical background. It contributes an integrated Agency Theory, TOE, and Institutional Theory framework. This study demonstrates the methodology for conducting bilingual content analysis for AI adoption proxies in the MENA region. The remainder of this study is organized as follows: Section \u003cspan refid=\"Sec2\" class=\"InternalRef\"\u003e2\u003c/span\u003e discusses the theoretical framework and hypotheses, and Section \u003cspan refid=\"Sec11\" class=\"InternalRef\"\u003e3\u003c/span\u003e explains the methodology. Section \u003cspan refid=\"Sec15\" class=\"InternalRef\"\u003e4\u003c/span\u003e presents the results, and Section \u003cspan refid=\"Sec21\" class=\"InternalRef\"\u003e5\u003c/span\u003e discusses the findings. Finally, Section \u003cspan refid=\"Sec25\" class=\"InternalRef\"\u003e6\u003c/span\u003e concludes the paper.\u003c/p\u003e"},{"header":"2. Theoretical Framework and Hypothesis","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Theoretical Foundations\u003c/h2\u003e \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e \u003ch2\u003e2.1.1 Agency Theory and Algorithmic Monitoring\u003c/h2\u003e \u003cp\u003eUltimately, the reason for audit efficiency concerns from the standpoint of Agency Theory and how AI technology could potentially help with it becomes clear. The disconnection between ownership and control results in information asymmetry in the relationship between principals and agents (Jensen \u0026amp; Meckling, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1976\u003c/span\u003e; Fama \u0026amp; Jensen, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e1983\u003c/span\u003e). External audits can act as a monitoring process to reduce the severity of this asymmetry. The possibility of full-population testing and temporal compression via near-real-time analysis are two mechanisms through which AI tools redefine audit IT monitoring (Munoko et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Lehner et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Tied to agency theory, the adoption of AI is expected to increase audit efficiency by decreasing the time and effort required to reach a certain level of assurance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.1.2 Technology-Organization-Environment (TOE) Framework\u003c/h2\u003e \u003cp\u003eAccording to the TOE framework, the successful adoption of technology is a function of technological readiness, organizational or operational capacity, and environmental influences such as need or pressure (Tornatzky \u0026amp; Fleischer, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; Marhraoui, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). In the organization dimension, outputs may be based on IT governance, which functions to \u0026ldquo;decide which AI projects get prioritized, resourced and evaluated among competing ones\u0026rdquo; (Almaqtari, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Emett, Eulerich, Pikoos, \u0026amp; Wood, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). Schiff, Kelley, and Ib\u0026aacute;\u0026ntilde;ez (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) determined that the use of AI by IT auditors has some technical limitations that can be managed through IT governance. It is based on the TOE in which IT governance is hypothesized as a moderator.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.1.3 Institutional Theory\u003c/h2\u003e \u003cp\u003eInstitutional Theory explains the exogenous factors encouraging AI adoption, including coercive elements relating to CBJ regulations, mimetic factors in terms of competitor imitation, and normative factors linked to professional standards (DiMaggio \u0026amp; Powell, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e1983\u003c/span\u003e; Scott, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Bonsu, Wang \u0026amp; Guo, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Roncato, Medeiros, and Lerner (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2026\u003c/span\u003e) found that the influence of regulatory obligations positively impacts the adoption and compliance quality of governance practices in emerging markets.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Literature Review\u003c/h2\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.2.1 AI Adoption and Audit Efficiency\u003c/h2\u003e \u003cp\u003eWhile the overall literature on the role of AI in auditing is still limited, as previously mentioned, some scholars have moved beyond conceptual frameworks to empirical analyses. For instance, Munoko, Buck, and Tumunya (2020) provided a seminal analysis of the opportunities and risks of AI in auditing. Based on 25,408 firm-year samples in China, Tan, Cheung, and Yu (2025) empirically measured how using AI in auditing improves audit quality and reduces audit lags. Rahman, Chowdhury, and Islam (2024) reported from their empirical study that when AI is used jointly, audit report lag is significantly reduced. In a similar context, Lai (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) empirically documented that AI reduces audit fees by reducing information asymmetry. Fawzi Khan, Hiba Jan, and Samina Zia-ul-haq (2025) found that AI adoption empowers integrated financial reporting in the Gulf Cooperation Council markets, and this relationship is mediated by audit quality. Through a controlled experiment conducted in Egypt, Abouelela et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) confirmed the role of AI in enhancing audit planning. Peng, Gong, and Zhong (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2026\u003c/span\u003e) reported how AI improves the quality of accounting information. Fajardo and Neiva (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) found in their empirical analysis that machine learning algorithms significantly improved auditing in the Brazilian government sector. Finally, Agostino et al. (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) confirm the vast shortage of empirical studies on algorithmic technologies in accounting.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2 IT Governance as a Moderating Mechanism\u003c/h2\u003e \u003cp\u003eAlmaqtari (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), surveying 228 Saudi IT professionals, identified IT governance as a significant predictor of the extent to which companies integrate AI into their operations. Hu, Xie, Cheng, Liu, and Zhang (2023) contend that AI governance is a key factor in the success of AI-powered auditing practices. Emett, Feldman, and Paim (2026) present a governance framework validated through interviews and focus groups that include 69 control considerations. Lacmanović and Skare (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) analyzed several companies and showed that auditing AI for bias is not a standardized process. Silic, Silic, and Kind-Tr\u0026uuml;ller (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) conceive shadow AI as an instance of governance failure. Neiroukh and Caglar (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) prove, through interviews with managers in Jordanian banks, that AI and IT governance positively impact the quality of information systems. Li and Goel (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) identify data governance and model explainability as critical requirements for AI auditability. von Zahn, Zacharias, Lowin, Chen, and Hinz (2025) provide a design framework for conformity assessments involving AI with components of fairness and explainability.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Hypothesis Development\u003c/h2\u003e \u003cp\u003e \u003cb\u003eH1: AI adoption is negatively associated with audit delay in Jordanian commercial banks, such that banks with higher AI adoption scores exhibit shorter audit delays, ceteris paribus\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eThe direct relationship is built from the monitoring function of Agency Theory and supported by Tan et al. (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e, Rahman et al. (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2024\u003c/span\u003e, Lai \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e, Peng et al. (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). If these AI tools enable data extraction, reconciliation, and confirmation matching, they are expected to reduce the audit process time.\u003c/p\u003e \u003cp\u003e \u003cb\u003eH2: IT governance maturity positively moderates the relationship between AI adoption and audit efficiency, such that the negative association between the AI adoption score and audit delay is stronger for banks with higher IT governance maturity\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eThe moderating effect stems from the TOE and has been supported by Almaqtari (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), Neiroukh and Caglar (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), and Hu et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Mature IT governance is correlated with a robust data infrastructure foundation and accountability that allow investments in AI to have a higher probability of leading to efficiency improvements.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Research Methodology","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Research Design and Sample\u003c/h2\u003e \u003cp\u003eThis study is guided by a positivist philosophy and should be understood as deductive in nature (Bryman, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The sample covers the entire population of all commercial and Islamic banks present in the ASE during the period from to 2015\u0026ndash;2024, which totals 15 banks each year, contributing to a balanced panel of 150 bank-year observations. The data comprise audited financial statements, corporate governance disclosures, and hand-collected audit reports.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Variable Measurement\u003c/h2\u003e \u003cp\u003e \u003cb\u003eDependent Variable: Audit Delay.\u003c/b\u003e It is the number of calendar days between the fiscal year-end (December 31) and the audit report, an audit timeliness proxy that is often related to audit efficiency (Knechel \u0026amp; Sharma, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Tan et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Rahman et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Abouelela et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cb\u003eThe independent Variable: AI Adoption Score.\u003c/b\u003e Developed by the researcher based on bilingual content analysis guided by Loughran and McDonald (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Machine learning, artificial intelligence, process automation, and data analytics. It was normalized for report length and scaled from [0,1]. This measure has been tested for measuring the intensity of technology adoption (Peng, Tong, and Xie, 2026; Xie, Peng, \u0026amp; Tong,2026; Li and Chen, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cb\u003eModerating Variable: IT Governance.\u003c/b\u003e A COBIT-based composite index (0\u0026ndash;1) from strategic IT alignment, IT risk management, IT resource management, IT performance measurement, IT compliance, IT value delivery, and data governance (Almaqtari, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Hu et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; ISACA, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cb\u003eControl Variables.\u003c/b\u003e Revenue growth represents the operational complexity (Knechel and Sharma \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Min et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Big4 accounts for audit firm resources and technology ( Abouelela et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Saif-Alyousfi, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2025\u003c/span\u003e ). ROA was used by Daugherty and Dickins (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) and Xie et al. (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2026\u003c/span\u003e) to proxy profitability and risk. Use represents the financial risk (Khan et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Nguyen, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Continuous variables are winsorized at the 1st/99th percentiles (Kothari, Leone, and Wasley, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eVariable Definitions and Measurement\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMeasurement\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRole\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSource\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI_Score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNormalized AI keyword frequency (0\u0026ndash;1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIndependent\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAnnual reports\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAudit_Delay\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDays: fiscal year-end to audit report\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDependent\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAudit reports\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIT_Gov\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCOBIT-aligned index (0\u0026ndash;1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eModerator\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGovernance filings\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrowth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYear-on-year revenue growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eControl\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFinancial statements\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBig4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1 if Big Four auditor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eControl\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAudit reports\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eROA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNet income / total assets\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eControl\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFinancial statements\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal liabilities / total assets\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eControl\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFinancial statements\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Econometric Specification\u003c/h2\u003e \u003cp\u003eModel 1 (Baseline): \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Audit\\_Delay\\:=\\:\\beta\\:₀\\:+\\:\\beta\\:₁AI\\_Score\\:+\\:\\beta\\:₂Growth\\:+\\:\\beta\\:₃Big4\\:+\\:\\beta\\:₄ROA\\:+\\:\\beta\\:₅Leverage\\:+\\:\\delta\\:ₜ\\:+\\:\\epsilon\\:\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eModel 2 (Direct): \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Audit\\_Delay\\:=\\:\\beta\\:₀\\:+\\:\\beta\\:₁AI\\_Score\\:+\\:\\beta\\:₂IT\\_Gov\\:+\\:\\gamma\\:{\\prime\\:}Controls\\:+\\:\\delta\\:ₜ\\:+\\:\\epsilon\\:\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eModel 3 (Interaction): \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Audit\\_Delay\\:=\\:\\beta\\:₀\\:+\\:\\beta\\:₁AI\\_Score\\:+\\:\\beta\\:₂IT\\_Gov\\:+\\:\\beta\\:₃(AI\\_Score\\times\\:IT\\_Gov)\\:+\\:\\gamma\\:{\\prime\\:}Controls\\:+\\:\\delta\\:ₜ\\:+\\:\\epsilon\\:\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eModel 4 (Robustness): specification with COVID dummy replacing year-fixed effects. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Audit\\_Delay\\:=\\:\\beta\\:₀\\:+\\:\\beta\\:₁AI\\_Score\\:+\\:\\beta\\:₂IT\\_Gov\\:+\\:\\beta\\:₃(AI\\_Score\\times\\:IT\\_Gov)\\:+\\:\\gamma\\:{\\prime\\:}Controls\\:+\\:COVIDₜ\\:+\\:\\epsilon\\:\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eTime-fixed effects year (δₜ) absorb time-varying shocks. The interaction term was mean-centered to reduce multicollinearity (Aiken \u0026amp; West, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1991\u003c/span\u003e). In case of detected cross-sectional dependence (Pesaran CD test), all models are estimated with Driscoll-Kraay standard errors, robust to heteroskedasticity, autocorrelation, and cross-sectional dependence (Driscoll \u0026amp; Kraay, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1998\u003c/span\u003e; Hoechle, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Vogelsang, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). However, with N\u0026thinsp;=\u0026thinsp;15 banks, it is not safe to use cluster-robust SE (Cameron et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). All analyses were performed using the PyCharm IDE and Python 3.12.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Empirical Results","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Descriptive Statistics\u003c/h2\u003e \u003cp\u003eThe descriptive statistics are presented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The average audit delay is 62.39 days (s.d. = 6.86) and ranges between 48 and 75 days, in accordance with the CBJ\u0026rsquo;s deadline of March 31. The mean adoption score is 0.178, with a standard deviation of 0.155, with banks reporting scores from virtually no AI adoption to 0.576 adoption, indicating moderate levels of adoption at the time of data collection but a general upward trend. The IT governance composite had a mean of 0.500 (SD\u0026thinsp;=\u0026thinsp;0.250) and ranged from 0.000 to 1.000, illustrating a high degree of variability. For controls, the percentage of bank-years that were Big Four audited was 71.3%, the average ROA was 5.02%, and average use was 34.41%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive Statistics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMedian\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMin\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eP25\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eP75\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eMax\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAudit Delay (days)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e62.3885\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6.8550\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e75\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI Adoption Score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.1783\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1271\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1549\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0068\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.0474\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.2776\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.5762\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIT Governance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.5000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.4810\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.2500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.3257\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.6532\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e1.0000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRevenue Growth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0920\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0891\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1192\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.1703\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.0069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.1816\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.3488\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBig 4 Auditor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.7133\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.4537\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1.0000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e1.0000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReturn on Assets\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.0502\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.0486\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.0396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.0396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.0220\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.0791\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.1377\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFinancial Leverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.3441\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.3360\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.1538\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.0500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.2268\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.4647\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.6952\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Correlation Analysis and Multicollinearity\u003c/h2\u003e \u003cp\u003eThe Pearson correlation coefficient with the variance inflation factor (VIF) is presented in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. AI adoption and audit delay have a significantly high negative correlation (\u0026minus;\u0026thinsp;0.765), supporting H1. IT governance is correlated with audit delay at a value of \u0026minus;\u0026thinsp;0.485. The highest correlation among the independent variables was between AI_Score and IT_Gov, with a coefficient of 0.303, which is below the accepted level of 0.70 (Kennedy, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). The VIF values lie between 1.02 and 1.12, indicating that there is no multicollinearity.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePearson Correlation Matrix with VIF\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAI_Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIT_Gov\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGrowth\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBig4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eROA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eLeverage\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eVIF\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAI_Score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.303***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIT_Gov\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.303***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.048\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.096\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.019\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrowth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.048\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.128\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBig4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.096\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.128\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.099\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.080\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eROA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.089\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.099\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.019\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.080\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Diagnostic Tests\u003c/h2\u003e \u003cp\u003eThe panel diagnostic tests are presented in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. The Breusch-Pagan (LM\u0026thinsp;=\u0026thinsp;13.73, p\u0026thinsp;=\u0026thinsp;0.546) and White tests (LM\u0026thinsp;=\u0026thinsp;73.82, p\u0026thinsp;=\u0026thinsp;0.877) do not reject the homoskedasticity. The Durbin-Watson statistic is 2.29, and the Wooldridge AR(1) test results F\u0026thinsp;=\u0026thinsp;2.83 (p\u0026thinsp;=\u0026thinsp;0.095), indicating mild autocorrelation. However, the Pesaran CD test rejects cross-sectional independence (CD\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;2.19, p\u0026thinsp;=\u0026thinsp;0.029), which is why Driscoll-Kraay standard errors are used.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePanel Diagnostic Tests\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTest\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStatistic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDecision\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBreusch-Pagan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13.733\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.546\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFail to reject H₀\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e73.818\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.877\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFail to reject H₀\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWooldridge AR(1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.832\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.095\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFail to reject H₀\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePesaran CD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;2.188\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eReject H₀ **\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Regression Results\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e reports the results of all four model specifications with Driscoll-Kraay standard errors.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRegression Results (Driscoll-Kraay Standard Errors)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModel 1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eModel 2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eModel 3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eModel 4 (Robust.)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eAI_Score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;26.220***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;26.780***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;26.071***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;24.275***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(1.801)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(2.120)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(2.006)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(2.200)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eIT_Gov\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;7.530***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;7.516***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;7.453***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.651)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.599)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.573)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eAI_Score \u0026times; IT_Gov\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;6.221**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;9.804***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(2.397)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(2.064)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eGrowth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;11.109***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;10.568***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;10.222***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;9.688***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(2.289)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(1.163)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(1.308)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(1.132)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eBig4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;3.143***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;3.604***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;3.591***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;3.382***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(0.435)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(0.460)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(0.451)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.435)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eROA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;23.134***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;21.569***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;21.365***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;21.948***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(3.353)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(5.012)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(4.815)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(4.391)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eLeverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.578***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.629***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9.723***\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9.420***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(2.174)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(1.692)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(1.658)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(1.675)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eCOVID\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;2.163***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(0.543)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear FE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.763\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.826\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.827\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.815\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdj. R\u0026sup2;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.807\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.807\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.805\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNotes\u003c/em\u003e: \u003cb\u003e*** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003c/b\u003e, \u003cb\u003e** p\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/b\u003e, \u003cb\u003e* p\u0026thinsp;\u0026lt;\u0026thinsp;0.10\u003c/b\u003e. \u003cem\u003eThe standard errors are in parentheses. Models1-3 include year effects, and Model 4 uses a COVID dummy (1 for 2020\u0026ndash;2024).\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eModel 1\u003c/b\u003e shows that AI adoption has a significant negative relationship with audit delay ( = -26.22, p\u0026thinsp;=\u0026thinsp;0.01). The increase in the AI score (0.155) of one standard deviation translates to approximately 4.06 less or a 6.5% better score compared to 62.39 days, which is the average. All controls are significant: Growth (β = \u0026minus;11.11, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), Big4 (β = \u0026minus;3.14, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), ROA (β = \u0026minus;23.13, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), and Leverage (β\u0026thinsp;=\u0026thinsp;9.58, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). R\u0026sup2; = 0.763.\u003c/p\u003e \u003cp\u003e \u003cb\u003eModel 2\u003c/b\u003e adds IT governance, substantially improving the fit(R 2\u0026thinsp;=\u0026thinsp;0.826). IT governance is inversely significant ( = -7.53, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), and AI adoption is not significantly important ( = -26.78, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), proving the independent impacts on audit efficiency.\u003c/p\u003e \u003cp\u003e \u003cb\u003eModel 3\u003c/b\u003e adds the interaction term, which is negative and substantial ( 3 = -6.22, p\u0026thinsp;=\u0026thinsp;0.011), which directly validates H2. It can be seen in a simple slopes analysis that the marginal AI effect for less than 24.52 at the low IT governance (\u0026minus;\u0026thinsp;1 SD) and above 27.63 at the high IT governance (+\u0026thinsp;1 SD) is strong.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eModel 4 (robustness)\u003c/strong\u003e \u003cp\u003ethe year FE is replaced by a COVID dummy. With a very large magnitude, the interaction is very relevant (β 3 = -9.80, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). COVID was noteworthy and negative (216 = -, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), which proves the acceleration of digitization during the pandemic.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.5 Hypothesis Testing Summary\u003c/h2\u003e \u003cp\u003e \u003cstrong\u003eH1 (Supported)\u003c/strong\u003e \u003cp\u003eAI adoption is significantly and negatively associated with audit delay across all four models (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), with coefficients ranging from \u0026minus;\u0026thinsp;24.28 to \u0026minus;\u0026thinsp;26.78. The adoption trajectory from near-zero (2015) to 0.41 (2024) implies approximately 10.7 fewer days, a 17% efficiency gain.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eH2 (Supported)\u003c/strong\u003e \u003cp\u003eThe interaction term is significant in Model 3 (β₃ = \u0026minus;6.22, p\u0026thinsp;=\u0026thinsp;0.011) and Model 4 (β₃ = \u0026minus;9.80, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Banks with high IT governance extract significantly greater efficiency gains from their AI investments.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"5. Discussion","content":"\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e5.1 AI Adoption and Audit Efficiency\u003c/h2\u003e \u003cp\u003eThe finding that AI adoption significantly reduces audit delay complements and contributes to the sparse existing literature. Our findings (β = \u0026minus;26) are comparable and consistent with those of Tan et al. (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), who examined the effect of AI adoption on audit lag in China and concluded that AI reduces audit lag. Our study is based on the context of Jordan, where AI adoption is still in its infancy; thus, an increase in adoption yields a marginal but significant improvement in the efficiency of accounting firms\u0026rsquo; operations as they shift from manual to automated processes. Our findings further support Bonsu et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), who reported that fintech adoption enhances accounting practices in emerging markets, and Peng et al. (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2026\u003c/span\u003e), who found that AI improves accounting information.\u003c/p\u003e \u003cp\u003eThis is particularly relevant given the audit technology environment in Jordanian banks. While audits in developed countries are established within a technological context, banks in emerging markets such as Jordan are in the process of shifting from a manual environment to a technological one, thereby creating more significant efficiency implications. Fajardo and Neiva (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) examine the efficiency effects of artificial intelligence in the audit records of the public sector in Brazil and find similar results.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e5.2 The Moderating Role of IT Governance\u003c/h2\u003e \u003cp\u003eThe significance of the interaction term indicates that IT governance enhances the efficiency of AI. The significance of this interaction term supports the organizational aspect of the TOE framework: organizations must do something with their technological capabilities to derive value. This finding supports Almaqtari\u0026rsquo;s (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) survey-based findings on IT value creation but advances this and other extant AI literature by using archival rather than perceptual data and by focusing on a specific value outcome instead of a generic measure. This finding is also consistent with Neiroukh and Caglar\u0026rsquo;s (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) study of Jordanian banks, which found that AI and governance positively affect the quality of information.\u003c/p\u003e \u003cp\u003eThe moderation plot shows that when IT governance is low, the slope for AI and audit delay is flatter, but it steeply increases when IT governance is high (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This result supports Hu et al.\u0026rsquo;s (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) finding that improving AI governance is a top priority and Emett et al.\u0026rsquo;s (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2026\u003c/span\u003e) argument that governance structures are needed to mitigate the risks of deploying AI. Both Lacmanović and Skare (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) and von Zahn et al. (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) reinforce this need for standardized governance frameworks.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e5.3 Control Variable Findings\u003c/h2\u003e \u003cp\u003eAll the control variables are statistically significant and expected. The negative relationship with growth may capture the operational dynamism that accompanies the growth phase (Min et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The impact of the Big Four on audit report lag (approximately 3.5 fewer days) is in line with superior technological infrastructure and supports Abouelela et al.\u0026rsquo;s (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) findings. Return on assets (ROA) negatively affects audit report lag, confirming that audits of profitable banks are less risky and require less effort (Xie et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). In contrast, use has a positive effect, indicating that audits of highly leveraged banks are a focus (Khan et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The dummy variable for the COVID pandemic is statistically significant, and this result verifies that the pandemic triggered digitization efforts (Albous et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"6. Conclusion","content":"\u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e6.1 Summary\u003c/h2\u003e \u003cp\u003eThis study empirically tests the impact of artificial intelligence (AI) adoption on audit efficiency and the moderating role of information technology (IT) governance in Jordanian banks. The analysis was based on 150 bank-year observations from all 15 banks listed on the Amman Stock Exchange (ASE) from 2015 to 2024. The results show that the adoption of AI significantly reduces audit delay; IT governance reduces audit delay and enhances AI\u0026rsquo;s efficiency effects; and the results are controlled for the COVID-19 period and several model modifications.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003e6.2 Theoretical Contributions\u003c/h2\u003e \u003cp\u003eTheoretically, this study contributes to the literature in three dimensions. First, this study adds to Agency Theory by extending it to algorithmic monitoring. The findings reveal that AI serves as an information source to assist in the governance process by collapsing the monitoring time element instead of providing situational information, as established in previous studies (Lehner et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Gorwa, Binns \u0026amp; Katzenbach, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Second, this study adds to the TOE framework by conceptualizing IT governance as a moderating organizational capability, not a direct antecedent, in the chain linking IT resources to organizational performance. Third, this study integrates Institutional Theory to explain the role of coercive, mimetic, and normative motives behind MENA banks\u0026rsquo; adoption and governance practices of AI technology (Roncato et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2026\u003c/span\u003e; Albous et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003e6.3 Practical Implications\u003c/h2\u003e \u003cp\u003eFor bank managers, AI investment will not bring efficiency benefits unless accompanied by a suitable investment in IT governance. For the CBJ, it may be wise to continue fostering AI adoption and governance maturity. For audit firms, clients\u0026rsquo; AI maturity and governance strength should be included as part of the overall engagement strategy. Finally, for MENA policymakers, structured regulatory encouragement can lead to positive AI adoption in the region, provided that governance issues are addressed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003e6.4 Limitations and Future Research\u003c/h2\u003e \u003cp\u003eThe study\u0026rsquo;s limitations are its relatively small sample size (15 banks), AI measurement based on content analysis, binary scoring of IT governance, and sole use of an efficiency proxy. Future research should consider a larger sample from the MENA region, use patent-based AI measures alongside textual-based measures (Ding \u0026amp; Gu, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2026\u003c/span\u003e), employ a more nuanced IT governance measure, and consider additional outcome variables such as audit fees and the probability of financial statement restatements. To firmly establish causality, an instrumental variable methodology based on specific regulatory events should be used.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003eData Availability\u003c/p\u003e\n\u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;The data used to support these conclusions are derived from the annual reports and financial statements of companies traded on the Amman Stock Exchange (ASE). The results and analyses of the datasets presented in the present study can be accessed from the authors of this study upon reasonable request.\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003ePrincess Nourah bint Abdulrahman University Researchers Supporting Project number (\u003cstrong\u003ePNURSP2026R861\u003c/strong\u003e), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.\u003c/p\u003e\n\u003cp\u003eCompeting Interests\u003c/p\u003e\n\u003cp\u003eThe authors agree that this research was conducted in the absence of any self-benefit or commercial or financial conflicts.\u003c/p\u003e\n\u003cp\u003eEthical Statements\u003c/p\u003e\n\u003cp\u003eThis study uses publicly available secondary data from annual and financial reports. No human participants were involved; therefore, ethical approval was not required.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbouelela O, Diab A, Saleh S (2025) The relationship between AI and audit planning in emerging economies. SAGE Open, 15(4)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAcemoglu D, Restrepo P (2019) Automation and new tasks are also important. J Economic Perspect 33(2):3\u0026ndash;30\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgostino D, Louren\u0026ccedil;o R, Jorge S, Bracci E, Cruz I (2025) Data science and public sector accounting. Public Money \u0026amp; Management\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAiken LS, West SG (1991) Multiple regression: Testing and interpreting interactions. Sage\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlbous MR, Stephens M, Al-Jayyousi OR (2025) AI and the GCC workforce. Humanit Social Sci Commun 12(1):1649\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlmaqtari FA (2024) IT governance in the integration of AI in accounting and auditing. Economies 12(8):199\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBao HX, Liu WF, Dai Z (2025) AI vs. public administrators. Technol Forecast Soc Chang 215:124102\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBonsu MOA, Wang Y, Guo YS (2023) Does Fintech lead to better accounting practices? Acc Res J 36(2/3):129\u0026ndash;147\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBryman A (2016) Social Research Methods, 5th edn. Oxford University Press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCameron AC, Gelbach JB, Miller DL (2008) Bootstrap-based improvements for inference with clustered errors. Rev Econ Stat 90(3):414\u0026ndash;427\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChai L et al (2025) AI-generated content and ESG. Romanian J Economic Forecast 28(3):5\u0026ndash;23\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrawford J, Nilsson F (2023) Integrating ESG risks into control and reporting. Handbook of Big Data and Analytics in Accounting and Auditing. Springer, pp 255\u0026ndash;277\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDaugherty B, Dickins D (2010) Perceptions of auditor independence and financial reporting quality. Adv Acc Behav Res 13:169\u0026ndash;194\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiMaggio PJ, Powell WW (1983) The iron cage revisited. Am Sociol Rev 48(2):147\u0026ndash;160\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing H, Gu HY (2026) Supply chain shared auditor and corporate AI adoption. Asia Pacific Journal of Marketing and Logistics\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDriscoll JC, Kraay AC (1998) Consistent covariance matrix estimation with spatially dependent panel data. Rev Econ Stat 80(4):549\u0026ndash;560\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEisikovits N, Johnson WC, Markelevich A (2025) Should accountants be afraid of AI? Acc Horizons 39(2):117\u0026ndash;123\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEmett S, Eulerich M, Pikoos J, Wood DA (2026) Generative AI governance framework. Accounting Horizons\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFajardo B, Neiva S (2025) AI application for auditing in the Brazilian federal government. Public Money \u0026amp; Management\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFama EF, Jensen MC (1983) Separation of ownership and control. J Law Econ 26(2):301\u0026ndash;325\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGambhir B, Srivastava A, Gupta N (2025) AI in audit and accounting: A qualitative exploration. Asian J Acc Gov, 24\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGorwa R, Binns R, Katzenbach C (2020) Algorithmic content moderation. Big Data Soc, 7(1)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarymawan I et al (2023) CEO facial masculinity and tax avoidance. Cogent Bus Manage 10(1):2171644\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHenderson MD (2025) Agentic AI and the ethics of leadership maintenance. Leadership \u0026amp; Organization Development Journal\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoechle D (2007) Robust standard errors for panel regressions with cross-sectional dependence. Stata J 7(3):281\u0026ndash;312\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu KH, Chen FH, Hsu MF, Tzeng GH (2023) Governance of AI in a business audit. Financial Innov 9(1):117\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eISACA (2019) COBIT 2019 Framework. ISACA\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIslam MS et al (2026) AI-enabled environmental auditing and sustainability disclosure. Corporate Social Responsibility and Environmental Management\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJensen MC, Meckling WH (1976) Theory of the firm. J Financ Econ 3(4):305\u0026ndash;360\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJlifi B et al (2026) Reconciling divergence among ESG scores. Computational Economics\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKennedy P (2008) A guide to econometrics, 6th edn. Blackwell\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhan F, Jan SU, Zia-ul-haq HM (2025) AI adoption, audit quality and integrated financial reporting in GCC. Asian Rev Acc 33(3):464\u0026ndash;495\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKnechel WR, Sharma DS (2012) Auditor-provided nonaudit services and audit effectiveness. Auditing: J Pract Theory 31(4):73\u0026ndash;105\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKothari SP, Leone AJ, Wasley CE (2005) Performance matched discretionary accrual measures. J Account Econ 39(1):163\u0026ndash;197\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLacmanović S, Skare M (2025) AI bias auditing. Rev Acc Finance 24(3):375\u0026ndash;400\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLai J (2025) AI applications and audit fees. Int Rev Econ Finance 103:104421\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLehner OM et al (2022) AI-based decision-making in accounting and auditing. Acc Auditing Account J 35(9):109\u0026ndash;135\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi GH, Chen YT (2025) Digitally powered ESG evaluation. Romanian J Economic Forecast 28(3):89\u0026ndash;109\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi N, Kim M, Dai J, Vasarhelyi MA (2024) Using AI in ESG assurance. J Emerg Technol Acc 21(2):83\u0026ndash;99\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi YQ, Goel S (2025) Bridging IT auditors and AI auditing. Adv Acc 69:100842\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin TLJ, Maginnis J (2025) Generative AI and edge AI in auditing. Current Issues in Auditing\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoughran T, McDonald B (2011) When is a liability not a liability? J Finance 66(1):35\u0026ndash;65\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarhraoui MA (2026) Drivers and outcomes of AI adoption. IEEE Trans Eng Manage 73:920\u0026ndash;935\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMin H, Mirza SS, Huang CM (2025) Digital transformation and audit opinions. Journal of Corporate Accounting and Finance\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMunoko I, Brown-Liburd HL, Vasarhelyi M (2020) Ethical implications of AI in auditing. J Bus Ethics 167(2):209\u0026ndash;234\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeiroukh N, Caglar D (2025) Information systems quality and corporate sustainability. Systems 13(7):537\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen TA (2025) Bibliometric analysis of ESG risk. Bus Strategy Dev, 8(3), e70163\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeng Z, Gong YT, Zhong X (2026) Can AI improve accounting information quality? Applied Economics\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRahman MJ, Ziru A (2023) Clients' digitalization and audit quality. Int J Acc Inform Manage 31(2):221\u0026ndash;246\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRahman MJ, Zhu HT, Yue L (2024) AI adoption, audit quality and efficiency. Managerial Auditing J 39(6):668\u0026ndash;699\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoncato S, Medeiros LM, Lerner AF (2026) Regulation and corporate governance practices. J Financial Regul Compliance 34(2):177\u0026ndash;197\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaif-Alyousfi AYH (2025) AI, information environment, and capital market efficiency. Res Int Bus Finance 79:103094\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchiff DS, Kelley S, Ib\u0026aacute;\u0026ntilde;ez JC (2024) AI ethics auditing. Big Data Soc, 11(4)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScott WR (2014) Institutions and organizations, 4th edn. Sage\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilic M, Silic D, Kind-Tr\u0026uuml;ller K (2025) From shadow IT to shadow AI. Strategic Change\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan JH, Chang SM, Zheng Y, Chan KC (2025) Client AI and audit quality. Int Rev Financial Anal 104:104271\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTornatzky LG, Fleischer M (1990) The processes of technological innovation. Lexington Books\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVogelsang TJ (2012) Heteroskedasticity, autocorrelation, and spatial correlation robust inference. J Econ 166(2):303\u0026ndash;319\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Zahn M et al (2025) Navigating AI conformity. Electron Markets 35(1):24\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXie LF, Peng ZB, Tong XG (2026) AI strategy, earnings management, and corporate fraud. Econ Model 156:107460\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao X, Tong YS, Lee H, Shahzad U (2025) AI in enhancing corporate environmental disclosure. Energy Econ 148:108680\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Audit Efficiency, AI Adoption, IT Governance, Driscoll-Kraay, Jordanian Banks, Panel Data","lastPublishedDoi":"10.21203/rs.3.rs-9348015/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9348015/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAI is rapidly developing in auditing practices worldwide, but there is little empirical evidence on the link between AI and audit efficiency among emerging market banks. In addition, the organizational variables through which AI improves audit efficiency— for example, IT governance maturity— are not theoretically and empirically theoretically sound and empirically untested. This study explores the effect of AI adoption on audit delay in Jordanian banks and the altering role of IT governance maturity. The data for 2015–2024 are summarized in a balanced panel of 15 commercial and Islamic banks listed on the ASE, indicating 150 bank years. The normalized score of AI adoption is obtained through bilingual content analysis of annual reports, and a CoBIT-aligned composite index for IT governance maturity is measured by a standardized score of the bilingual content analysis of the annual reports. Four model specifications employ pooled OLS regression with Driscoll–Kraay standard errors robust to heteroskedasticity, autocorrelation, and cross-sectional dependence. This suggests that AI adoption significantly reduces audit delay (= 26.07, p = 0.01); an increased AI score = 4.06 fewer audit days, which translates into a 6.5% efficiency improvement; and an increase in one standard deviation in the AI score yields 4.06 fewer audit days, or a 6.5% improvement in accuracy. I conclude that the interaction between AI adoption and IT governance maturity is negative and significant: 3 = 6.22, p = 0.011, and banks with larger IT governance maturity significantly increase the efficiency of audit performance. The results are reliable when the year fixed effects are replaced with a COVID-19 control variable. The present study contributes the first archival evidence from emerging-market banks in the MENA region on the AI–audit efficiency nexus, incorporates Agency Theory, TOE and Institutional Theory, and emphasizes IT governance as a critical barrier to achieving an efficient use of AI in auditing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eJEL Classification:\u003c/strong\u003e M41, M42, G21, G34, O33\u003c/p\u003e","manuscriptTitle":"Algorithmic Governance and Audit Efficiency: The Role of AI Adoption and IT Governance in Jordanian Commercial Banks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-13 08:04:30","doi":"10.21203/rs.3.rs-9348015/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-16T14:49:45+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-15T17:50:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-14T04:11:32+00:00","index":"","fulltext":""},{"type":"submitted","content":"Humanities and Social Sciences Communications","date":"2026-04-07T17:02:16+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"28baf36e-156f-4473-8f52-9b960dc6ff32","owner":[],"postedDate":"April 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":65882159,"name":"Business and commerce/Business and management"},{"id":65882160,"name":"Social science/Business and management"},{"id":65882161,"name":"Business and commerce/Information systems and information technology"},{"id":65882162,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-03T11:38:17+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-13 08:04:30","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9348015","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9348015","identity":"rs-9348015","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}