Adapting Code Review Processes to the Conditions of AI-Assisted Software Development

Yurii Bezhentsev

Yurii Bezhentsev Software Developer Houston, Texas, United States

Keywords: code review, AI-assisted software development, code generation, development stability, cognitive load, risk management, validation processes

Abstract

The article presents an analysis of the transformation of code review processes in the context of software development with the use of artificial intelligence. The study is conducted in the format of a systematic review and analytical synthesis of academic publications focusing on code generation, development productivity, and the organization of change validation processes. The primary focus is on the relationship between the speed of code creation, the characteristics of introduced changes, and the ability of the review process to ensure their reliable interpretation and safe integration. The key parameters determining the effectiveness of code review are examined, including temporal characteristics, the structure of changes, and cognitive load. It is established that the impact of code generation tools is indirect and manifests through the formation of asymmetry between the intensity of generation and the capacity for review. It is shown that even with improvements in the quality of automatically generated code, the risk of uncritical acceptance and the integration of hidden logical inconsistencies remains. An original adaptive code review model is proposed, reflecting a transition from single-stage review to a multi-level system that includes preliminary filtering, structured analysis, risk assessment, and a feedback loop. The results obtained allow code review to be considered as a controllable risk regulation mechanism determining the stability and predictability of the development process. The practical significance of the study lies in substantiating the need to structure code review processes, limit the scope of changes, and increase transparency in the origin of code fragments. The article will be useful for software developers, engineering managers, and specialists in software process management.

References

1. Adalsteinsson, F. S., Magnússon, B. B., Milicevic, M., Davidsson, A. N., & Cheng, C.-H. (2025). Rethinking code review workflows with LLM assistance: An empirical study. arXiv. https://doi.org/10.48550/arXiv.2505.16339
2. Alami, A., & Ernst, N. A. (2025). Human and machine: How software engineers perceive and engage with AI-assisted code reviews compared to their peers. arXiv. https://doi.org/10.48550/arXiv.2501.02092
3. Almeida, Y., Albuquerque, D., Dantas Filho, E., Muniz, F., de Farias Santos, K., Perkusich, M., Almeida, H., & Perkusich, A. (2024). AICodeReview: Advancing code quality with AI-enhanced reviews. SoftwareX. https://doi.org/10.1016/j.softx.2024.101677
4. Bistarelli, S., Fiore, M., Mercanti, I., et al. (2025). Usage of large language model for code generation tasks: A review. SN Computer Science, 6, 673. https://doi.org/10.1007/s42979-025-04241-5
5. Celik, A., & Mahmoud, Q. H. (2025). A review of large language models for automated test case generation. Machine Learning and Knowledge Extraction, 7(3), 97. https://doi.org/10.3390/make7030097
6. Gullstrand Heander, L., Söderberg, E., & Rydenfält, C. (2026). Code review as decision-making: Building a cognitive model from the questions asked during code review. Empirical Software Engineering, 31, 54. https://doi.org/10.1007/s10664-025-10791-2
7. Heander, L., Söderberg, E., & Rydenfält, C. (2025). Support, not automation: Towards AI-supported code review for code quality and beyond. In Proceedings of the 33rd ACM International Conference on the Foundations of Software Engineering (FSE Companion '25) (pp. 591–595). Association for Computing Machinery. https://doi.org/10.1145/3696630.3728505
8. İçöz, B., & Biricik, G. (2026). Context-aware code review automation: A retrieval-augmented approach. Applied Sciences, 16(4), 1875. https://doi.org/10.3390/app16041875
9. Kansab, S., Bordeleau, F., & Tizghadam, A. (2025). RevMine: An LLM-assisted tool for code review mining and analysis across Git platforms. arXiv. https://doi.org/10.48550/arXiv.2510.04796
10. Khade, P. S., & Sambhe, R. U. (2025). Artificial intelligence in software development: A review of code generation, testing, maintenance and security. International Journal of Current Science Research and Review, 8(4), 1632–1641. https://doi.org/10.47191/ijcsrr/V8-i4-08
11. Khare, V., Saini, V., Sharma, D., Kumar, A., Rana, A., & Yadav, A. (2025). DeputyDev—AI powered developer assistant: Breaking the code review logjam through contextual AI to boost developer productivity. arXiv. https://doi.org/10.48550/arXiv.2508.09676
12. Mahfoodh, H., Hammad, M., Alqaralleh, B. A. Y., & Zreikat, A. I. (2026). Evaluating LLMs for source code generation and summarization using machine learning classification and ranking. Computers, 15(2), 119. https://doi.org/10.3390/computers15020119
13. Maru, G. G., Lee, S., Ji, S., Ko, S.-K., & Im, H. (2025). Neural methods for programming: A comprehensive survey and future directions. Applied Sciences, 15(22), 12150. https://doi.org/10.3390/app152212150
14. Mavridou, E., Vrochidou, E., Kalampokas, T., Kanakaris, V., & Papakostas, G. A. (2025). AI-powered software development: A systematic review of recommender systems for programmers. Computers, 14(4), 119. https://doi.org/10.3390/computers14040119
15. Pahl, C., Sezen, Ö. C., & Hofer, F. (2026). Artificial intelligence for infrastructure-as-code—A systematic literature review. Electronics, 15(4), 755. https://doi.org/10.3390/electronics15040755