Towards an Appropriate Level of Reliance on AI: A Preliminary Reliance-Control Framework for AI in Software Engineering
Samuel Ferino, Rashina Hoda, John Grundy, Christoph Treude
Why It Matters
What makes this one worth your time
This paper is significant as it addresses the critical issue of balancing AI reliance, which can impact productivity and cognitive skills in software development, providing a foundation for responsible AI tool usage.
A framework to balance reliance on AI tools in software engineering is proposed.
Summary
The paper introduces a preliminary reliance-control framework to address the balance between overreliance and underreliance on AI tools, particularly Large Language Models, in software engineering. Through interviews with developers, it explores how varying levels of control can optimize the use of AI, aiming to enhance productivity while maintaining critical thinking skills.
Key contributions
- Proposes a reliance-control framework for AI tool usage in software engineering.
Notable insights
- The level of control over AI tools can serve as an indicator of appropriate reliance, helping to avoid both overreliance and underreliance.
Possible limitations
- The framework is preliminary and based on a limited number of interviews, requiring further validation and exploration.
Abstract
arXiv:2604.10530v1 Announce Type: cross Abstract: How software developers interact with Artificial Intelligence (AI)-powered tools, including Large Language Models (LLMs), plays a vital role in how these AI-powered tools impact them. While overreliance on AI may lead to long-term negative consequences (e.g., atrophy of critical thinking skills); underreliance might deprive software developers of potential gains in productivity and quality. Based on twenty-two interviews with software developers on using LLMs for software development, we propose a preliminary reliance-control framework where the level of control can be used as a way to identify AI overreliance and underreliance. We also use it to recommend future research to further explore the different control levels supported by the current and emergent LLM-driven tools. Our paper contributes to the emerging discourse on AI overreliance and provides an understanding of the appropriate degree of reliance as essential to developers making the most of these powerful technologies. Our findings can help practitioners, educators, and policymakers promote responsible and effective use of AI tools.