ISSSN (Online): 3049-3684
Volume: 2 Issue: 1 | Open Access |12 Feb 2026
Review on AI in Coding - Future and Challenges
Prashanti Vibha, Engineering Student, T. R. Abhilashi Memorial Institute of Engineering and Technology
Abstract: Artificial intelligence (AI) is increasingly integrated into software development workflows, including code generation and automated testing. Major progress has occurred with large language models (LLMs) and specialized code models, but significant challenges remain — encompassing correctness, security, licensing, evaluation, and human factors. This paper reviews representative systems and organizes current research challenges, concluding with a forward-looking roadmap for future work.
Keywords: code generation, large language models, software engineering, evaluation, security, human–AI interaction
References
Feng, Z., Guo, D., Tang, D., Duan, N., Feng, X., Ding, M., & Jiang, Y. (2020). CodeBERT: A Pre-Trained Model for Programming and Natural Languages. arXiv preprint arXiv:2002.08155.
Li, Y., Choi, D., Chung, J., Kushman, N., … & Vinyals, O. (2022). Competition-Level Code Generation with AlphaCode. arXiv preprint arXiv:2203.07814.
Pearce, H., Sandoval, G., Nys, T., Karri, R., Garg, S., & Dolan-Gavitt, B. (2023). A User-Centered Security Evaluation of Copilot. ACM Digital Library.
Sandoval, G., Pearce, H., et al. (2023). Lost at C: A User Study on the Security Implications of Large Language Model Code Assistants. USENIX.
Zhang, H., et al. (2024). Deep Learning for Code Generation: A Survey. SciChina Information Sciences.
Pearce, H., et al. (2022). Asleep at the Keyboard? Assessing the Security of GitHub Copilot. IEEE Proceedings.