From Code Author to Code Reviewer: A Systematic Review of Practitioner Experience with Generative AI in Software Engineering
Abstract
Practitioner use of generative AI in software engineering has progressed faster than the evidence assessing its effects. This review examined 58 empirical studies published from 2024 to early 2026 on GitHub Copilot, ChatGPT, Claude, and Gemini. Using PRISMA-guided selection from five digital libraries, the studies addressed productivity, role shifts, code quality, adoption behaviour, and evidence gaps. Findings were synthesised into the AI Integration Consequence Model, a conceptual four-layer framework linking adoption conditions, role shifts, quality exposure, and organisational consequences. Evidence indicates that developers increasingly move from code production to code review, as more than half of AI-generated code is discarded after evaluation. Generative AI improves efficiency for boilerplate code, documentation, and unit-test scaffolding, but provides limited benefits for architectural decisions and enterprise debugging. A meta-analysis of 47 studies reported a 45.6-point gap between self-reported and objectively measured productivity. AI-generated code also showed exploitable vulnerabilities at estimated rates of 40–50%, varying by task, tool version, programming language, and measurement method. Social influence strongly predicted adoption intent, while concerns about peer judgement encouraged undisclosed use. Experienced engineers generally received an AI Boost, whereas junior engineers faced an AI Drag. Structured mentorship reduced this gap in one 14-month cohort study. Key evidence gaps include cross-cultural adoption, longitudinal skill development, and unreplicated gender disparities.
Related
- Sound and music in video games: A systematic review of player engagement and a meta-analysis of positively valenced player experience
- Difference-in-Differences Designs: A Practitioner’s Guide
- Causal Inference with Generative Artificial Intelligence: Application to Texts as Treatments
- Generative modeling via hierarchical tensor sketching
- Generative SDEs with Trainable Diffusion and Density-Based Losses
BibTeX
@article{isong2026from,
title = {{From Code Author to Code Reviewer: A Systematic Review of Practitioner Experience with Generative AI in Software Engineering}},
author = {Bassey Isong and Tshipuke Vahangwele and Nosipho Dladlu},
journal = {Journal of Information Systems and Informatics},
year = {2026},
month = oct,
doi = {10.63158/journalisi.v8i5.1826},
url = {https://doi.org/10.63158/journalisi.v8i5.1826},
note = {CC BY},
}
RIS
TY - JOUR
TI - From Code Author to Code Reviewer: A Systematic Review of Practitioner Experience with Generative AI in Software Engineering
AU - Bassey Isong
AU - Tshipuke Vahangwele
AU - Nosipho Dladlu
PY - 2026
DA - 2026/10/24
JO - Journal of Information Systems and Informatics
DO - 10.63158/journalisi.v8i5.1826
UR - https://doi.org/10.63158/journalisi.v8i5.1826
AB - Practitioner use of generative AI in software engineering has progressed faster than the evidence assessing its effects. This review examined 58 empirical studies published from 2024 to early 2026 on GitHub Copilot, ChatGPT, Claude, and Gemini. Using PRISMA-guided selection from five digital libraries, the studies addressed productivity, role shifts, code quality, adoption behaviour, and evidence gaps. Findings were synthesised into the AI Integration Consequence Model, a conceptual four-layer framework linking adoption conditions, role shifts, quality exposure, and organisational consequences. Evidence indicates that developers increasingly move from code production to code review, as more than half of AI-generated code is discarded after evaluation. Generative AI improves efficiency for boilerplate code, documentation, and unit-test scaffolding, but provides limited benefits for architectural decisions and enterprise debugging. A meta-analysis of 47 studies reported a 45.6-point gap between self-reported and objectively measured productivity. AI-generated code also showed exploitable vulnerabilities at estimated rates of 40–50%, varying by task, tool version, programming language, and measurement method. Social influence strongly predicted adoption intent, while concerns about peer judgement encouraged undisclosed use. Experienced engineers generally received an AI Boost, whereas junior engineers faced an AI Drag. Structured mentorship reduced this gap in one 14-month cohort study. Key evidence gaps include cross-cultural adoption, longitudinal skill development, and unreplicated gender disparities.
SN - 2656-5935
SN - 2656-4882
ER -