AI Adoption, Reporting Quality, and Romanian Listed Firms' Market Value: Artificial Neural Networks Approach

Authors

  • Mansour Alraja Northumbria University, Newcastle Business School (United Kingdom)
  • Billal Chikhi M'hamed Bougara Boumerdes University (Algeria)

Keywords:

Artificial Intelligence Adoption, Financial Reporting Quality, Market Value, Bucharest Stock Exchange, Discretionary Accruals, Multi-layer Perceptron

Abstract

This study examines the impact of corporate Artificial Intelligence (AI) adoption on Financial Reporting Quality (FRQ) and Market Value for technology firms listed on the Bucharest Stock Exchange (2020–2024). Using 150 firm-year observations, the research pairs linear Ordinary Least Squares (OLS) regressions with a non-linear Multi-layer Artificial Neural Network (ANN). OLS results indicate that higher AI disclosure improves FRQ by reducing discretionary accruals, while both AI adoption and enhanced FRQ positively signal market value (Tobin’s Q). Crucially, the ANN framework radically outpaces the linear model (R² = 0.91 vs. 0.64), proving that technological innovation valuation is fundamentally non-linear. Feature importance profiling identified AI disclosure as the primary market value predictor (28.5% weight). Consequently, AI integration acts as a potent corporate governance tool and value driver, justifying regulatory calls for standardized digital disclosure mandates.

References

Brynjolfsson, E., & McAfee, A. (2017). Machine, platform, crowd:

Harnessing our digital future. W.W. Norton & Company.

Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.

Healy, P. M., & Wahlen, J. M. (1999). A review of earnings management

literature and its implications for standard setting. Accounting Horizons,

(4), 365–383. https://doi.org/10.2308/acch.1999.13.4.365

Jensen, M. C. (2001). Value maximization, stakeholder theory, and the

corporate objective function. Journal of Applied Corporate Finance, 14(3),

–21. https://doi.org/10.1111/j.1745-6622.2001.tb00434.x

Kothari, S. P., Leone, A. J., & Wasley, C. E. (2005). Performance

matched discretionary accrual measures. Journal of Accounting and

Economics, 39(1), 163–197. https://doi.org/10.1016/j.jacceco.2004.11.002

Lev, B., & Gu, F. (2016). The end of accounting and the path forward for

investors and managers. John Wiley & Sons.

Schumaker, R. P., & Chen, H. (2009). Textual analysis of stock market

prediction using financial news articles. ACM Transactions on Information

Systems, 27(2), 1–19. https://doi.org/10.1145/1462198.1462204

Sutton, S. G., Holt, M., & Arnold, V. (2016). The reports of my death are

greatly exaggerated—Artificial intelligence research in accounting.

International Journal of Accounting Information Systems, 22, 60–73.

https://doi.org/10.1016/j.accinf.2016.07.005

Published

2026-09-01