AI-Driven Financial Transparency and Corporate Governance: Enhancing Accounting Practices with Evidence from Jordan

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors examine the impact of AI-driven financial transparency on corporate governance and regulatory reform. Using the method of stratified random sampling, it is found that AI significantly improves the effectiveness of corporate governance. The research is interesting and also reveals flaws in AI. Papers may be considered for publication after revision.

1. The summary should have quantitative results to support the conclusion.
2. There are too many keywords and they need to be cut.
3. The introduction raises five questions. Should the corresponding conclusion answer these five questions? The description of the paper framework is too simple and needs to be properly expanded. In addition, consider making a paper frame diagram.
4. A complete version of the questionnaire should be attached to the paper.
5. Many tables present the results without in-depth analysis of the meaning behind the data.
6. Figure 1 is not clear and needs to be redrawn.
7. Is the ranking of the results important or is the numerical value more important? Because some of the values are not very different from each other. Whether it can be classified and analyzed.
8. Recommendations and Further Studies after the conclusion, whether the content of these two parts is too much. Is it appropriate to put in the conclusion section. Consider putting it in the discussion section.

Author Response

For research article: AI-Driven Financial Transparency and Corporate Governance: Enhancing Accounting Practices with Evidence from Jordan.

We sincerely appreciate your thoughtful and constructive feedback on our manuscript. Your comments have been instrumental in improving the clarity, rigor, and depth of our study. Below, we provide a point-by-point response to your suggestions and outline the corresponding revisions made to the manuscript.

1. Abstract Revision

Comment: The summary should have quantitative results to support the conclusion.
Response: We have revised the abstract to include key quantitative findings, such as R² values for corporate governance effectiveness (0.582), risk management (0.502), and stakeholder engagement (0.681). These results provide empirical support for the study’s conclusions.

2. Keyword Reduction

Comment: There are too many keywords, and they need to be cut.
Response: We have refined the keyword list to focus on the most relevant terms: Artificial Intelligence, Corporate Governance, Financial Transparency, Risk Management, and Regulatory Compliance.

3. Introduction and Conclusion Alignment

Comment: The introduction raises five questions. Should the corresponding conclusion answer these five questions?
Response: We have explicitly addressed each research question in the conclusion, summarizing the key findings for AI’s impact on decision-making, risk management, transparency, stakeholder engagement, and corporate governance mechanisms.

4. Expansion of Paper Framework Section

Comment: The description of the paper framework is too simple and needs to be properly expanded. Consider making a paper frame diagram.
Response: The paper structure has been expanded, and a flow diagram illustrating the research framework has been included. The revised structure is:

• Section 1: Introduction (Research objectives, questions, and significance)
• Section 2: Literature Review (AI's role in governance and transparency)
• Section 3: Methodology (Stratified sampling, data collection, and analysis techniques)
• Section 4: Results (Quantitative findings with statistical significance)
• Section 5: Discussion (Interpretation, implications, and challenges)
• Section 6: Conclusion (Summary of key findings)

5. Questionnaire Inclusion

Comment: A complete version of the questionnaire should be attached to the paper.
Response: We have added the full questionnaire as Appendix A to enhance transparency in the data collection process.

6. In-Depth Analysis of Results

Comment: Many tables present the results without in-depth analysis of the meaning behind the data.
Response: We have expanded the results section with detailed interpretations of the findings, emphasizing their implications for corporate governance. Examples include AI’s role in executive decision-making (R = 0.763, p < 0.001) and its influence on risk management (R = 0.709, p < 0.001).

7. Figure 1 Clarity

Comment: Figure 1 is not clear and needs to be redrawn.
Response: Figure 1 has been redrawn with higher resolution and improved labeling for better readability.

8. Clarification on Rankings vs. Numerical Values

Comment: Is the ranking of the results important, or is the numerical value more important? Because some of the values are not very different from each other. Can they be classified and analyzed?
Response: We have categorized the results into high-impact, moderate-impact, and low-impact to improve clarity. A table has been added to show classification criteria based on statistical differences.

9. Reorganization of Recommendations and Further Studies

Comment: The Recommendations and Further Studies sections after the conclusion—whether the content of these two parts is too much. Is it appropriate to put it in the conclusion section? Consider putting it in the discussion section.
Response: The Recommendations and Further Studies sections have been relocated to the Discussion section to streamline the conclusion.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1. The study employs stratified random sampling without any theoretical or statistical justification for the distribution across sectors. The selection of 564 corporate professionals across industries is not explained, and the method does not seek to avoid selection bias, and therefore the findings are not dependable. Moreover, The conclusions just focus on the known benefits of AI for improving financial transparency and governance without providing any novel theoretical insights. The manuscript analysed general statements from last literature instead of conducting a critical analysis of the limitations or alternative perspectives.


2. While the research makes use of structural equation modeling (SEM) and multiple regression analyses, no evaluation of model assumptions such as normality, multicollinearity, or heteroscedasticity is presented. The high Cronbach's alpha values (above 0.97) suggest redundancy instead of validity, pointing to weaknesses in survey design.


3. Although the paper touches on bias in AI algorithms, data privacy concerns, and accountability concerns, it does not suggest any mitigation or how AI governance can be implemented in Jordan's corporate and legal landscapes. The discourse is theoretical with no policy recommendations. furthermore, The paper has weak theoretical grounding, not anchoring its findings in traditional governance theories like Agency Theory or Institutional Theory. The study largely presents correlations without causality testing.

Comments on the Quality of English Language

Fix all grammatical errors. 

Author Response

We sincerely appreciate your thoughtful and constructive feedback on our manuscript. Your comments have been instrumental in improving the clarity, rigor, and depth of our study. Below, we provide a point-by-point response to your suggestions and outline the corresponding revisions made to the manuscript.

Comment 1: The study employs stratified random sampling without any theoretical or statistical justification for the distribution across sectors. The selection of 564 corporate professionals across industries is not explained, and the method does not seek to avoid selection bias, making the findings less dependable. Additionally, the conclusions focus on known AI benefits rather than offering novel theoretical insights. The manuscript analyzes general statements from past literature instead of conducting a critical analysis of limitations or alternative perspectives.

Response:
We appreciate this valuable feedback and have made the following revisions to strengthen the methodological justification and theoretical contributions:

1. Justification for Stratified Random Sampling: We have revised the methodology section to explain the rationale for using stratified sampling. The sample was proportionally distributed across sectors (e.g., banking, manufacturing, and technology) based on their relative market representation in Jordan. A detailed justification for sectoral distribution has been added, citing relevant statistical and theoretical frameworks.
2. Avoidance of Selection Bias: To enhance the reliability of our findings, we have elaborated on the sample selection process. We now describe the steps taken to minimize selection bias, including the randomization procedure and inclusion criteria to ensure representativeness.
3. Novel Theoretical Insights: In the discussion section, we have expanded our theoretical contributions by introducing a critical perspective on AI-driven transparency. We highlight potential risks, ethical dilemmas, and regulatory challenges that were previously underexplored.
4. Critical Analysis of Literature: The literature review has been revised to incorporate alternative perspectives, limitations, and gaps in existing research. Instead of summarizing prior studies, we now provide a critical assessment of unresolved issues and conflicting viewpoints.

Comment 2: While the research makes use of structural equation modeling (SEM) and multiple regression analyses, no evaluation of model assumptions such as normality, multicollinearity, or heteroscedasticity is presented. The high Cronbach's alpha values (above 0.97) suggest redundancy instead of validity, pointing to weaknesses in survey design.

Response:
We appreciate the reviewer’s insightful feedback regarding the evaluation of model assumptions and the validity of our survey instrument. To address these concerns, we have made the following revisions and additions:

1. Skewness and Kurtosis Analysis

The skewness values range from -1.429 to -0.820, and kurtosis values range from -0.225 to 1.608. While some variables show mild to moderate departures from normality (e.g., AIIMPACT and TRANSPARENCY, which have kurtosis values above 1), these values remain within the acceptable range for large sample sizes (N = 564). Given the Central Limit Theorem, minor deviations from normality are unlikely to significantly impact the validity of parametric analyses.

2. Normality Tests (Kolmogorov-Smirnov and Shapiro-Wilk)

The Kolmogorov-Smirnov (K-S) and Shapiro-Wilk tests were conducted to assess the normality of the data. The results indicate that all variables deviate significantly from a normal distribution (p-values < 0.05). However, these tests tend to be overly sensitive in large samples (N = 564), often detecting even trivial deviations. Therefore, normality was further evaluated using skewness and kurtosis measures.

3. Homoscedasticity Assessment (Residual Scatter Plot)

A scatter plot of residuals against predicted values was generated to examine homoscedasticity. The visual inspection of the plot does not show a clear funnel shape or systematic pattern, indicating that the assumption of homoscedasticity holds. This suggests that residual variances remain consistent across all levels of predicted values.

Comment 3: Although the paper touches on bias in AI algorithms, data privacy concerns, and accountability concerns, it does not suggest any mitigation or how AI governance can be implemented in Jordan's corporate and legal landscapes. The discourse is theoretical with no policy recommendations. Furthermore, the paper has weak theoretical grounding, not anchoring its findings in traditional governance theories like Agency Theory or Institutional Theory. The study largely presents correlations without causality testing.

Response:
We sincerely appreciate your valuable feedback, which has guided us in strengthening the theoretical and practical implications of our study. In response to your concerns, we have made the following modifications:

1. AI Governance and Policy Recommendations
   We have expanded our discussion to include concrete policy recommendations for AI governance in Jordan’s corporate and legal landscape. Specifically, we propose the following measures:
   • Regulatory Frameworks: The adoption of AI-specific governance regulations aligned with international best practices (e.g., GDPR, OECD AI Principles).
   • Ethical AI Standards: Establishing sector-specific ethical guidelines to ensure AI transparency, accountability, and fairness.
   • Corporate AI Governance Committees: Encouraging firms to form AI governance committees to oversee AI adoption, ensuring compliance with financial and legal standards.
   • Stakeholder Engagement: Developing mechanisms for continuous monitoring and stakeholder participation in AI-driven decision-making.

These recommendations aim to address bias in AI algorithms, data privacy risks, and accountability concerns, ensuring that AI adoption aligns with Jordan’s regulatory and corporate governance landscape.

2. Strengthening Theoretical Grounding
   To enhance the theoretical foundation, we have explicitly anchored our study in two key governance theories:
   • Agency Theory (Jensen & Meckling, 1976): AI-driven financial transparency can reduce information asymmetry and mitigate agency conflicts between managers and shareholders. We discuss how AI-powered accounting practices enhance managerial accountability.
   • Institutional Theory (DiMaggio & Powell, 1983): Organizations in Jordan’s financial sector are adopting AI to conform to institutional pressures, such as regulatory requirements and industry best practices. We elaborate on how institutional norms shape AI adoption.

By integrating these theories, we strengthen our study’s conceptual underpinnings, providing a clearer framework for understanding AI’s role in corporate governance.

3. Causality Testing and Endogeneity
   While we acknowledge the importance of causality testing, our study primarily focuses on associational relationships rather than causal inferences. Conducting instrumental variable regression (IVR) or two-stage least squares (2SLS) requires strong instrumental variables, which are often difficult to identify in non-experimental settings. Given the nature of our dataset, we recognize this as a limitation and suggest future studies explore causal mechanisms using advanced econometric techniques.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Referee Report for
Sustainability

AI-Driven Financial Transparency and Corporate Governance: Enhancing

Accounting Practices with Evidence from Jordan

1. Summary

The authors study the effectiveness of AI-driven tools for financial transparency and corporate governance by stratified random sampling methods.

2. General Comments

Overall, this research could be meaningful. However, it suffers from substantial problems. I report just the typical as follows:

• This paper lacks theoretical originality and looks like industry reports. For instance, the authors do not mention “originality” or “theoretical (theory)” in the abstract.
  At least in the abstract, the authors could briefly review the literature and emphasize research gap. However, they do not.
• The authors heavily anchor this research on Al-Wasleh, but the authors shall sufficiently justify that Al-Wasleh could achieve AI-Driven financial transparency. Instead, the authors simply justify or just describe Al-Wasleh as follows:
• Firstly and the most importantly, the description does not focus on financial transparency (as the key term in the title) at all.
• Secondly, the authors do not provide academic reference, and (Al-Wasleh, 2024) is just an self-claimed corporate description at most.
  
  Lastly, the authors should analyze the target industries of Al-Wasleh.

Comments for author File: Comments.pdf

Author Response

For research article: AI-Driven Financial Transparency and Corporate Governance: Enhancing Accounting Practices with Evidence from Jordan.

We sincerely appreciate your thoughtful and constructive feedback on our manuscript. Your comments have been instrumental in improving the clarity, rigor, and depth of our study. Below, we provide a point-by-point response to your suggestions and outline the corresponding revisions made to the manuscript.

General Comments

1. Lack of Theoretical Originality and Resemblance to Industry Reports

Comment: This paper lacks theoretical originality and looks like an industry report. The authors do not mention "originality" or "theoretical (theory)" in the abstract.

Response: To enhance the theoretical contribution, we have revised the abstract to explicitly state the originality of our research and its theoretical implications. We now emphasize how our study extends corporate governance literature by integrating AI-driven financial transparency within the context of Institutional Theory and Agency Theory. Additionally, we clarify the theoretical underpinnings that differentiate our work from industry reports.

Revision: The abstract now highlights the theoretical framework used and the research gap it addresses, ensuring a stronger academic positioning.

2. Lack of Literature Review and Research Gap in the Abstract

Comment: At least in the abstract, the authors could briefly review the literature and emphasize the research gap. However, they do not.

Response: We have revised the abstract to include a brief literature review and a clear statement of the research gap. The revised version now acknowledges prior studies on AI’s role in corporate governance and financial transparency while underscoring the lack of empirical research linking AI-driven accounting practices to governance frameworks, particularly in emerging economies like Jordan.

Revision: The updated abstract explicitly states the research gap and the study’s contribution to filling this gap.

3. Over-Reliance on Al-Wasleh Without Justification

Comment: The authors heavily anchor this research on Al-Wasleh but do not sufficiently justify that Al-Wasleh could achieve AI-driven financial transparency. The description does not focus on financial transparency (as the key term in the title) at all.

Response: We acknowledge this concern and have addressed it in multiple ways:

1. Refined Justification: We have strengthened the justification for selecting Al-Wasleh by providing empirical evidence of its AI-driven financial practices. This includes examples of AI-based risk assessment, automated financial reporting, and fraud detection measures that contribute to financial transparency.
2. Expanded Theoretical Context: We have integrated Al-Wasleh’s AI-driven financial model into our discussion of Agency Theory and Institutional Theory, demonstrating how these technologies mitigate information asymmetry and enhance governance mechanisms.
3. Incorporation of Additional Case Studies: To avoid over-reliance on Al-Wasleh, we have briefly referenced other AI-driven financial transparency initiatives in Jordan and globally, ensuring a more balanced discussion.

Revision: The manuscript now provides a stronger theoretical and empirical justification for discussing Al-Wasleh in the context of financial transparency.

4. Lack of Academic References for Al-Wasleh

Comment: The authors do not provide academic references, and (Al-Wasleh, 2024) is just a self-claimed corporate description at most.

Response: We have addressed this issue by:

• Citing peer-reviewed articles and industry reports that discuss AI-driven financial transparency in similar organizations.
• Expanding the discussion to include AI adoption in financial governance frameworks beyond Al-Wasleh.
• Clarifying that Al-Wasleh serves as an illustrative example rather than the sole basis of our empirical findings.

Revision: The revised manuscript now incorporates academic references supporting AI-driven financial transparency and governance.

5. Lack of Industry Analysis for Al-Wasleh

Comment: The authors should analyze the target industries of Al-Wasleh.

Response: We have added a detailed industry analysis of Al-Wasleh, outlining the sectors it serves, such as banking, fintech, and regulatory technology (RegTech). This analysis highlights how AI applications in these industries enhance financial transparency and governance.

Revision: The revised paper includes a new subsection on industry analysis, ensuring a comprehensive discussion of Al-Wasleh’s role in financial governance.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Accepted

Author Response

For research article: AI-Driven Financial Transparency and Corporate Governance: Enhancing Accounting Practices with Evidence from Jordan.

We sincerely appreciate your thoughtful and constructive feedback on our manuscript. Your comments have been instrumental in improving the clarity, rigor, and depth of our study.

Thank you.

Best Regards

Prof Osama Shaban

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The authors substantially improve the paper.

Author Response

For research article: AI-Driven Financial Transparency and Corporate Governance: Enhancing Accounting Practices with Evidence from Jordan.

We sincerely appreciate your thoughtful and constructive feedback on our manuscript. Your comments have been instrumental in improving the clarity, rigor, and depth of our study.

Thank you.

Best Regards

Prof Osama Shaban

Author Response File: Author Response.docx