Abstract:With the advances in machine learning, there is a growing interest in AI-enabled tools for autocompleting source code. GitHub Copilot has been trained on billions of lines of open source GitHub code, and is one of such tools that has been increasingly used since its launch in June 2021. However, little effort has been devoted to understanding the practices, challenges, and expected features of using Copilot in programming for auto-completed source code from the point of view of practitioners. To this end, we conducted an empirical study by collecting and analyzing the data from Stack Overflow (SO) and GitHub Discussions. We searched and manually collected 303 SO posts and 927 GitHub discussions related to the usage of Copilot. We identified the programming languages, Integrated Development Environments (IDEs), technologies used with Copilot, functions implemented, benefits, limitations, and challenges when using Copilot. The results show that when practitioners use Copilot: (1) The major programming languages used with Copilot are JavaScript and Python, (2) the main IDE used with Copilot is Visual Studio Code, (3) the most common used technology with Copilot is <a class="link-external link-http" href="http://Node.js" rel="external noopener nofollow">this http URL</a>, (4) the leading function implemented by Copilot is data processing, (5) the main purpose of users using Copilot is to help generate code, (6) the significant benefit of using Copilot is useful code generation, (7) the main limitation encountered by practitioners when using Copilot is difficulty of integration, and (8) the most common expected feature is that Copilot can be integrated with more IDEs. Our results suggest that using Copilot is like a double-edged sword, which requires developers to carefully consider various aspects when deciding whether or not to use it. Our study provides empirically grounded foundations that could inform developers and practitioners, as well as provide a basis for future investigations.

GitHub Copilot: The Disrupting AI Companion Transforming the Developer Role and Application Lifecycle Management

Transforming Software Development: Evaluating the Efficiency and Challenges of GitHub Copilot in Real-World Projects

POTENCIALES Y DESAFÍOS DE GITHUB COPILOT COMO HERRAMIENTA DE INTELIGENCIA ARTIFICIAL

Demystifying Practices, Challenges and Expected Features of Using GitHub Copilot

Practices and Challenges of Using GitHub Copilot: An Empirical Study

Sea Change in Software Development: Economic and Productivity Analysis of the AI-Powered Developer Lifecycle

Grounded Copilot: How Programmers Interact with Code-Generating Models

Exploring the Problems, their Causes and Solutions of AI Pair Programming: A Study on GitHub and Stack Overflow

An Empirical Evaluation of GitHub Copilot's Code Suggestions

The Impact of AI on Developer Productivity: Evidence from GitHub Copilot

The Impact of Generative AI on Collaborative Open-Source Software Development: Evidence from GitHub Copilot

The Impact of AI Tool on Engineering at ANZ Bank An Empirical Study on GitHub Copilot within Corporate Environment

Navigating Applications Development in Generative AI

Copilot for Xcode: Exploring AI-Assisted Programming by Prompting Cloud-based Large Language Models

Assessing the Use of GitHub Copilot on Students of Engineering of Information Systems

Conversing with Copilot: Exploring Prompt Engineering for Solving CS1 Problems Using Natural Language

Generative AI for Pull Request Descriptions: Adoption, Impact, and Developer Interventions

The Productivity Effects of Generative AI: Evidence from a Field Experiment with GitHub Copilot

From Today's Code to Tomorrow's Symphony: The AI Transformation of Developer's Routine by 2030

From Copilot to Pilot: Towards AI Supported Software Development

Is GitHub's Copilot as Bad as Humans at Introducing Vulnerabilities in Code?