The Future of Software Engineering in an AI-Driven World
Valerio Terragni, Partha Roop, Kelly Blincoe
TL;DR
The paper tackles how AI-driven tools, particularly LLMs, can transform software engineering by enabling a symbiotic human-AI partnership across the full software lifecycle. It proposes a framework with an orchestrated set of AI subsystems that assist in requirements, design, development and testing, and maintenance, while preserving human oversight. Key contributions include identifying research challenges (e.g., requirements understanding, explainable design, AI-assisted testing with automated oracles, metamorphic testing, and maintenance decision-making) and proposing practical concepts like a prompt-friendly requirement language and APIzation of AI-generated artifacts. The study highlights the potential productivity gains and the need for careful handling of security, bias, and ethics, arguing for multi-disciplinary collaboration to realize the five-year vision.
Abstract
A paradigm shift is underway in Software Engineering, with AI systems such as LLMs gaining increasing importance for improving software development productivity. This trend is anticipated to persist. In the next five years, we will likely see an increasing symbiotic partnership between human developers and AI. The Software Engineering research community cannot afford to overlook this trend; we must address the key research challenges posed by the integration of AI into the software development process. In this paper, we present our vision of the future of software development in an AI-Driven world and explore the key challenges that our research community should address to realize this vision.