Artificial Intelligence in Software Engineering: A Holistic Review of Models, Tools, and Future Directions

Brojo Kishore Mishra

¹Department of Computer Science & Engineering, NIST University, Berhampur, Odisha, India

Email: [email protected]

²Department of Computer Science & Engineering, NIST University, Berhampur, Odisha, India

Email: [email protected]

³Department of Computer Science & Engineering, NIST University, Berhampur, Odisha, India

Email: [email protected]

⁴Department of Computer Science & Engineering, NIST University, Berhampur, Odisha, India

Email: [email protected]

*Corresponding author

Artificial Intelligence (AI) has increasingly become integral to software engineering (SE), yet much of the existing research, including the foundational review by Kokol (2023), tends to focus on broad categorizations without thoroughly examining the profound effects of emerging AI technologies. This paper extends that foundational analysis by presenting an in-depth, multi-faceted overview of recent advances in AI-driven software engineering. It highlights critical but often overlooked areas such as Explainable AI (XAI), Large Language Models (LLMs), AI-enhanced DevOps, and the ethical challenges associated with intelligent systems. Furthermore, the study surveys practical AI applications throughout the software development lifecycle—from automated testing to deployment forecasting—while pinpointing persistent research gaps and technical obstacles. By synthesizing insights from various disciplines and the latest empirical findings, this work provides a strategic framework aimed at fostering trustworthy, scalable, and sustainable AI integration in software engineering. The contribution deepens previous syntheses and equips researchers and industry professionals with clear guidance for leveraging AI in the future of software development.

Artificial Intelligence (AI), Software Engineering, Explainable AI (XAI), Large Language Models (LLMs), DevOps Automation, Ethical AI, AI-Augmented Development, Software Testing, Code Generation, Software Lifecycle, Trustworthy AI, Sustainable AI, Knowledge Synthesis

Santosh Kumar Kar, Brojo Kishore Mishra, Arvind Kumar Sahu, Sanjit Kumar Acharya (2026). Artificial Intelligence in Software Engineering: A Holistic Review of Models, Tools, and Future Directions. Journal of Artificial Intelligence and Systems, 8, 18–43. https://doi.org/10.33969/AIS.2026.080102.

[1] Kokol, P. (2023). The Use of AI in Software Engineering: A Synthetic Knowledge Synthesis of the Recent Research Literature. Information, 14(3), 148. https://doi.org/10.3390/info14030148.

[2] Arora, L., et al. (2025). Explainable Artificial Intelligence Techniques for Software Development Lifecycle: A Phase-specific Survey. arXiv:2505.07058.

[3] Mohammadkhani, A.H., et al. (2023). A Systematic Literature Review of Explainable AI for Software Engineering. arXiv:2302.06065.

[4] Tantithamthavorn, C., et al. (2020). Explainable AI for Software Engineering. arXiv:2012.01614.

[5] Liu, A., & Chi, V. (2024). From Defects to Demands: A Unified LLM-Based Framework for Software Repair and Requirements. arXiv:2412.05098.

[6] Tufano, M., et al. (2022). Next-Gen Software Engineering: AI-Assisted Big Models. arXiv:2409.18048.

[7] Guo, J., et al. (2020). CodeBERT: A Pre-Trained Model for Programming and Natural Languages. arXiv:2002.08155.

[8] Zhang, Y., et al. (2021). GraphCodeBERT: Pre-training Code Representations with Data Flow. arXiv:2009.08366.

[9] Ma, Z. (2024). Current Applications and Future Prospects of AI in Software Engineering. Advances in Engineering Innovation, 13, 71–75.

[10] Geng, X., et al. (2023). Comment Generation with LLMs for Diverse Developer Intents. Proceedings of EMNLP 2023.

[11] Amalfitano, D., et al. (2023). Artificial Intelligence Applied to Software Testing: A Tertiary Study. ACM Computing Surveys, 56(3), 3616372.

[12] Zhao, Y., et al. (2015). Test Case Prioritization Based on Source Code Changes. IEEE ICSME.

[13] Harman, M., et al. (2012). Search-Based Software Engineering: Trends and Techniques. ACM Computing Surveys, 45(1), 11.

[14] Mohan, M., & Greer, D. (2019). Many-Objective Automated Refactoring. Information and Software Technology, 111, 1–15.

[15] Tatineni, S., & Mustyala, A. (2021). Machine Learning in DevOps for Release Management. Journal of Systems and Software, 180, 111014.

[16] Zhang, H., et al. (2020). Predicting Deployment Failures in CI Pipelines Using ML. IEEE ICSME.

[17] Vakkuri, V., et al. (2020). Ethical Considerations in AI Courses. ACM ITiCSE.

[18] Winfield, A.F., et al. (2019). Ethical Principles for Robotics and AI. Science and Engineering Ethics, 25(4), 1037–1054.

[19] Kulkarni, V., et al. (2022). Intelligent Software Engineering with AI. In ISDA 2022 (pp. 67–82). Springer.

[20] Batarseh, F.A., et al. (2020). The Application of AI in Software Engineering. In Data Democracy (pp. 179–232). Academic Press.

[21] Wei, Z., et al. (2022). CLEAR: Contrastive Learning-Based API Recommendation. IEEE/ACM ICSE 2022.

[22] Zhang, X., et al. (2023). ToolCoder: API Selection for Code Generation. ACM ISSTA 2023.

[23] Patil, S., et al. (2023). Gorilla: A Fine-Tuned LLaMA-Based Model for API Verification. EMNLP 2023.

[24] Mastropaolo, E., et al. (2023). Comment Completion Using T5. IEEE ICSME 2023.

[25] Lo, D. (2023). Trustworthy and Synergistic AI for SE: Vision and Roadmap. arXiv:2309.04142.

[26] Ozkaya, I. (2023). AI-Augmented Software Development Processes. IEEE Software, 40(1), 4–9.

[27] Belzner, L., et al. (2024). LLM-Assisted Software Engineering: Challenges and a Case Study. In AI and Reality, Springer.

[28] Sofian, H., et al. (2022). Systematic Mapping of AI Techniques in SE. IEEE Access, 10, 51021–51040.

[29] Daun, M., & Brings, J. (2023). How ChatGPT Will Change Software Engineering Education. ACM ITiCSE 2023.

[30] Majumdar, S., et al. (2023). Generative AI for Software Metadata: FIRE 2023 Track Overview. FIRE 2023 Proceedings.

[31] Cao, S., et al. (2024). SLR on Explainability in ML/DL-Based SE. arXiv:2401.14617.

[32] Završnik, J., et al. (2024). AI and Pediatrics: A Synthetic Knowledge Synthesis Approach. Electronics, 13, 512.

[33] Satpute, R.S., & Agrawal, A. (2023). Pragmatic Ambiguity in NL Requirements. IJSAAE, 11, 249–259.

[34] Geng, X., et al. (2023). LLM-Based Comment Generation for Developer Intents. EMNLP 2023.

[35] Lo, D. (2023). Trustworthy and Synergistic AI for Software Engineering: Vision and Roadmaps. arXiv:2309.04142.

[36] Zhang, X., et al. (2023). CodeXGLUE: A Benchmark Dataset and Evaluation Toolkit for Code Intelligence Tasks. arXiv:2102.04664.