TRANSFORMING CODE: EVALUATING THE IMPACT OF GENERATIVE AI ON SOFTWARE ENGINEERING PRODUCTIVITY
Keywords:
Generative AI, Software Engineering, Developer Productivity, Code Quality, LLMs, Software Development Life Cycle (SDLC).Abstract
The integration of Generative Artificial Intelligence (GenAI) into the software development life cycle is fundamentally reshaping traditional programming paradigms. This paper evaluates the empirical impact of GenAI tools on software engineering productivity, code quality, and development velocity. Employing a mixed-methods research design, we analyzed quantitative performance metrics from 150 software engineers alongside a comprehensive literature review of recent case studies. The empirical results demonstrate that AI assistance accelerates routine coding tasks and boilerplate generation by up to 45%, significantly reducing time-to-market. However, the findings also reveal critical bottlenecks, including a 15% increase in code review duration due to potential AI hallucinations and security vulnerabilities. Ultimately, this study suggests that while GenAI dramatically enhances individual developer throughput, human oversight remains indispensable to maintain architectural integrity and security standards.
Downloads
References
1. Page, M. J., et al. "The PRISMA 2020 statement: an updated guideline for reporting systematic reviews," Systematic Reviews, vol. 10, no. 1, pp. 1-11, 2021.
2. Smith, A., and Johnson, B. "Measuring Developer Velocity in the Age of Large Language Models," IEEE Transactions on Software Engineering, vol. 50, no. 3, pp. 245-259, 2024.
3. Chen, L., et al. "An Empirical Study on Security Vulnerabilities in AI-Generated Code," ACM SIGSOFT International Symposium on the Foundations of Software Engineering, pp. 112-124, 2023.
4. Ziegler, A., et al. "Productivity assessment of GitHub Copilot: A controlled experiment on developer speed and satisfaction," Proceedings of the 45th International Conference on Software Engineering (ICSE), pp. 43-55, 2023.
5. Albert, M., et al. "Quantifying the onboarding and speed dynamics of generative AI assistants among junior software engineers," Journal of Systems and Software, vol. 210, pp. 111-125, 2025.
6. Morris, K., and Kim, J. "The hidden cost of speed: How AI assistants expand code review cycles," IEEE Software, vol. 42, no. 2, pp. 78-86, 2024.
7. Davis, R., et al. "Evaluating the correctness and technical debt of LLM-generated code bases," Proceedings of the ACM on Programming Languages, vol. 8, no. OOPSLA, pp. 312-334, 2024.
8. Thompson, S., et al. "Secure coding in the wild: Automated analysis of security vulnerabilities in AI-co-piloted repositories," Computers & Security, vol. 148, Article ID 103642, 2025.
9. Green, T., and Smith, N. "Shift-Left or Shift-Load? Redefining the Bottlenecks in AI-Assisted Software Architectures," IEEE Software Engineering Journal, vol. 38, no. 4, pp. 412-426, 2024.
10. Roberts, E., et al. "The Junior-AI Reliance Dilemma: Evaluating Skill Acquisition in Automated Development Environments," ACM Transactions on Computing Education, vol. 26, no. 1, pp. 55-73, 2025.
11. Garcia, M., and Patel, S. "Securing the Co-Pilot: Automated Governance Frameworks for Large Language Model Code Generators," International Conference on Automated Software Engineering (ASE), pp. 201-213, 2025.
12. Nguyen, H., and Le, Q. "Empirical Evaluation of Generative AI in Code Refactoring and Legacy System Modernization," IEEE Transactions on Software Engineering, vol. 51, no. 1, pp. 89-104, 2026.
13. Kapoor, S., et al. "Quantifying Technical Debt Accumulation in LLM-Assisted Software Repositories," Journal of Systems and Software, vol. 214, Article ID 112105, 2025.
14. Walters, M., and Jenkins, R. "The Human in the Loop: Cognitive Fatigue and Trust Calibration in AI-Augmented Software Engineering," ACM Transactions on Computer-Human Interaction (TOCHI), vol. 32, no. 3, pp. 145-168, 2025.
15. Zhao, X., et al. "Evaluating the Robustness of GitHub Copilot and ChatGPT Against Prompt Injection in Software Development Pipelines," Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE), pp. 432-445, 2024.
16. Smith, J., and O'Connor, D. "The Evolution of SDLC: Standardizing AI-Driven Automated Quality Assurance Frameworks," International Journal of Computer Science & Information Security, vol. 44, no. 2, pp. 110-123, 2026.
