Copilot Impact Studies: Measuring Productivity, Trust, and Skill Evolution in Enterprise Developer Teams
DOI:
https://doi.org/10.22399/ijcesen.4217Keywords:
Copilot Impact, AI, AI-driven coding assistantsAbstract
The incorporation of artificial intelligence-driven coding assistants in enterprise software development is a paradigmatic shift in the way development teams think and implement technical solutions. This article examines the many-sided effects of AI copilot technology on developer productivity, trust calibration, and skill development in different enterprise contexts. By using a mixed-methods paradigm that integrates quantitative performance data with qualitative data derived from developer experiences, findings show a multifaceted reality where productivity improvements occur in an uneven pattern across task types and levels of experience. Although routine implementation tasks reveal hn≤icated that subjects who had been given an AI assistant tended to generate incorrect and insecure solutions to cryptography problems, with those who were provided with the assistant writing substantially less secure code (p = 0.05) and tending to be more confident in their insecure solutions (p < 0.001) [2]. Additionally, the enterprise environment provides special considerations related to security, compliance, and intellectual property that distinguish it from open-source or startup environments on which much of the current research has been performed.
References
[1] Sida Peng et al., "The Impact of AI on Developer Productivity: Evidence from GitHub Copilot", arXiv, 2023. [Online]. Available: https://arxiv.org/pdf/2302.06590
[2] Neil Perry et al., "Do Users Write More Insecure Code with AI Assistants?", arXiv, 2023. [Online]. Available: https://arxiv.org/pdf/2211.03622
[3] Albert Ziegler et al., "Productivity Assessment of Neural Code Completion", arXiv, 2022. [Online]. Available: https://arxiv.org/pdf/2205.06537
[4] Alisa Welter et al., "From Developer Pairs to AI Copilots: A Comparative Study on Knowledge Transfer", arXiv, Jun. 2025. [Online]. Available: https://arxiv.org/pdf/2506.04785
[5] Mohamed Soliman et al., "Mining software repositories for software architecture — A systematic mapping study", ScienceDirect, May 2025. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0950584925000163
[6] Patrice Seuwou et al., "User Acceptance of Information Technology: A Critical Review of Technology Acceptance Models and the Decision to Invest in Information Security", ResearchGate, 2016. [Online]. Available: https://www.researchgate.net/publication/312265224_User_Acceptance_of_Information_Technology_A_Critical_Review_of_Technology_Acceptance_Models_and_the_Decision_to_Invest_in_Information_Security
[7] Shraddha Barke et al., "Grounded Copilot: How Programmers Interact with Code-Generating Models", arXiv, 2022. [Online]. Available: https://arxiv.org/pdf/2206.15000
[8] Jakub Res et al., "Enhancing Security of AI-Based Code Synthesis with GitHub Copilot via Cheap and Efficient Prompt-Engineering", arXiv/Semantic Scholar, 2024. [Online]. Available: https://www.semanticscholar.org/reader/707d50923a9c758bd06eccc30efcb83352fccfd4
[9] Danie Smit et al., "The impact of GitHub Copilot on developer productivity from a software engineering body of knowledge perspective", ResearchGate, 2024. [Online]. Available: https://www.researchgate.net/publication/381609417_The_impact_of_GitHub_Copilot_on_developer_productivity_from_a_software_engineering_body_of_knowledge_perspective
[10] Gaurav Rohatgi, "Unlocking Developer Productivity: A Deep Dive into GitHub Copilot’s AI-Powered Code Completion", IJERT, 2024. [Online]. Available: https://www.ijert.org/unlocking-developer-productivity-a-deep-dive-into-github-copilots-ai-powered-code-completion
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
