Enhancing Software Engineering with AI: Key Insights from ChatGPT
Abstract
Artificial intelligence (AI) is currently a prominent topic in the field of software engineering. AI has greatly transformed software engineering by providing advanced tools that may boost the effectiveness and efficiency of various stages of the system development life cycle. In this paper, we examine the role of AI by focusing primarily on ChatGPT, an OpenAI language model. The research explores how ChatGPT can assist in software engineering tasks such as code generation, bug fixing, and documentation development. Through an analysis of real-world scenarios, the case study highlights the benefits and challenges of using AI tools in these three areas. The findings show that although AI can accelerate the output and simplify processes, its application must be carefully considered because it has issues in terms of lack of context awareness regarding performance considerations, project-specific requirements, its inability to access real-time logs or inspect environments, and its tendency to produce documentation that doesn't capture the necessary details for complex systems.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The finding of no statistically significant difference after adopting Copilot differs from prior studies (Al-Ahmad et al., 2024; Cui et al., 2024; Peng et al., 2023; Rajbhoj et al., 2024).
“For Copilot users, our findings suggest a relatively small increase in average weekly activity; however, there is no statistically significant difference after adopting Copilot, which differs from prior studies (Al-Ahmad et al., 2024; Cui et al., 2024; Peng et al., 2023; Rajbhoj et al., 2024).”
From Developer Productivity With and Without GitHub Copilot: A Longitudinal Mixed-Methods Case Study (2026) · cites this paper as Al-Ahmad et al., 2024
Controlled lab and field experiments on GenAI productivity often fail to capture the complex and nuanced process of real-world software development.
“Some of the studies on developer productivity and GenAI use apply controlled lab settings (Al-Ahmad et al., 2024) or controlled field experiments (Cui et al., 2024). These are great for studying causal effects, but often fail to encapsulate the complex and nuanced process of a real-world software development setting (Brynjolfsson et al., 2025).”
From Developer Productivity With and Without GitHub Copilot: A Longitudinal Mixed-Methods Case Study (2026) · cites this paper as Al-Ahmad et al., 2024
Prior studies focus on short-term gains in controlled or semi-controlled settings, either by measuring task completion time or by simulating software projects.
“Instead, they focus on short-term gains in controlled or semi-controlled settings, either by measuring task completion time (Peng et al., 2023), or by simulating software projects to illustrate how GenAI tools can assist developers during different phases of the development process (Al-Ahmad et al., 2024; Rajbhoj et al., 2024).”
From Developer Productivity With and Without GitHub Copilot: A Longitudinal Mixed-Methods Case Study (2026) · cites this paper as Al-Ahmad et al., 2024
Cites
AI revolutionizing industries worldwide: A comprehensive overview of its diverse applications
Rashid, A. B., & Kausik, M. A. K. (2024) · Hybrid Advances · doi:10.1016/j.hybadv.2024.100277 · Cited by 764 · cited 1× · see citing works
“In recent years, artificial intelligence has profoundly influenced nearly every sector, including software engineering [7, 8].”
Ghimire, P., Kim, K., & Acharya, M. (2024) · Buildings · doi:10.3390/buildings14010220 · Cited by 175 · cited 1× · see citing works
“However, AI-generated documentation does not fully capture the details of specific projects, so ongoing human input is needed to improve it [29, 30].”
‘Human oversight’ in the EU artificial intelligence act: what, when and by whom?
Enqvist, L. (2023) · Law, Innovation and Technology · doi:10.1080/17579961.2023.2245683 · Cited by 73 · cited 1× · see citing works
“While AI holds tremendous potential for advancing software engineering, certain questions emerge concerning quality and the level of human oversight required [10, 11].”
Flagged first, then by citations
Cited by
Sorted by citations
Developer Productivity With and Without GitHub Copilot: A Longitudinal Mixed-Methods Case Study
Stray, V., Brandtzæg, E. G., Wivestad, V., Barbala, A., & Moe, N. B. (2026) · Proceedings of the Annual Hawaii International Conference on System Sciences · doi:10.24251/hicss.2026.880 · Cited by 9 · see citing works
“For Copilot users, our findings suggest a relatively small increase in average weekly activity; however, there is no statistically significant difference after adopting Copilot, which differs from prior studies (Al-Ahmad et al., 2024; Cui et al., 2024; Peng et al., 2023; Rajbhoj et al., 2024).”
Influential
- Authors
- Al-Ahmad, A., Kahtan, H., Tahat, L., & Tahat, T.
- Affiliations
- Gulf University for Science and Technology,Business Technology Center,Mishref,Kuwait
- Cardiff Metropolitan University (GB)
- Gulf University for Science & Technology (KW)
- Peace Corps (US)
- Venue
- 2024 International Conference on Decision Aid Sciences and Applications (DASA)
- Publisher
- IEEE
- Published
- 2024-12-11
- Pages
- 1-5
- OpenAlex
- W4406522157
- Cited by
- 4
- References
- 3
- Citations by year
- 2026: 1 · 2025: 3
- FWCI
- 1.52
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International Conference on Decision Aid Sciences and Applications (DASA) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- Physical Sciences › Computer Science › Information Systems › Software Engineering Techniques and Practices
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence Applications
- Keywords
- ChatGPT, large language models, artificial intelligence in software development, code generation, bug fixes, software documentation, system development life cycle, context awareness
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
