Predicting Developer Acceptance of AI-Generated Code Suggestions
Jing Jiang, Liehao Li, Jinyun Hou, Xin Tan, Li Zhang
TL;DR
This work addresses the high interruption rate of AI-generated code suggestions by quantitatively identifying drivers of acceptance. Leveraging 66,329 industrial developer–AI interactions, it demonstrates that developer habits and project preferences are strong predictors of acceptance, while in-situ signals are less informative at generation time. The authors introduce CSAP, a lightweight neural predictor that integrates multi-source features to decide acceptance before display, achieving up to 0.973 accuracy on imbalanced data and 0.922 on balanced data, and outperforming production and LLM baselines. The results advocate for post-generation, personalized filtering and emphasize the importance of fine-grained interaction telemetry to tailor AI-assisted programming, addressing model drift and tool-chain evolution in real-world deployments.
Abstract
AI-assisted programming tools are widely adopted, yet their practical utility is often undermined by undesired suggestions that interrupt developer workflows and cause frustration. While existing research has explored developer-AI interactions when programming qualitatively, a significant gap remains in quantitative analysis of developers' acceptance of AI-generated code suggestions, partly because the necessary fine-grained interaction data is often proprietary. To bridge this gap, this paper conducts an empirical study using 66,329 industrial developer-AI interactions from a large technology company. We analyze features that are significantly different between accepted code suggestions and rejected ones. We find that accepted suggestions are characterized by significantly higher historical acceptance counts and ratios for both developers and projects, longer generation intervals, shorter preceding code context in the project, and older IDE versions. Based on these findings, we introduce CSAP (Code Suggestion Acceptance Prediction) to predict whether a developer will accept the code suggestion before it is displayed. Our evaluation of CSAP shows that it achieves the accuracy of 0.973 and 0.922 on imbalanced and balanced dataset respectively. Compared to a large language model baseline and an in-production industrial filter, CSAP relatively improves the accuracy by 12.6\% and 69.5\% on imbalanced dataset, and improves the accuracy by 87.0\% and 140.1\% on balanced dataset. Our results demonstrate that targeted personalization is a powerful approach for filtering out code suggestions with predicted rejection and reduce developer interruption. To the best of our knowledge, it is the first quantitative study of code suggestion acceptance on large-scale industrial data, and this work also sheds light on an important research direction of AI-assisted programming.