The impact of AI on software work: A systematic framework for understanding work transformation

Introduction
Software has become essential in modern society, with almost every field, from healthcare and finance to education and manufacturing, relying on software development. Globally, software work has been one of the fastest-growing professions, driving innovation and competitiveness. However, advanced AI technologies, particularly generative and language models, challenge this growth, fundamentally reshaping the nature of work. Work in the digital age is transforming at an unprecedented pace. It is about what tasks are performed and how they are executed, who is involved, and why work is done. AI seems to be becoming an active co-worker, collaborating with human professionals. This shift could redefine job roles and alter collaboration patterns: routine tasks are automated, creative and strategic functions are prioritized, and the boundaries between human expertise and machine efficiency are blurring. As work cycles accelerate, software professionals face both the promise of increased efficiency and the pressure to continuously adapt, learn, and manage complex ethical and organizational dilemmas. Understanding these multifaceted transformations requires a systemic perspective that connects everyday work practices with broader organizational and institutional dynamics. Our study develops a conceptual framework to address this complexity, guided by the research question: How can we systemically approach work transformation in the context of AI-changing software work?
Methods
The framework was developed in two phases: (1) a literature review on software work, AI’s role, and its challenges, and (2) an exploration of theories that help analyze the social dynamics of these changes. The selected theories were aimed to provide complementary perspectives, enabling a holistic understanding of AI’s impact on work and practical conclusions.
Results
We propose an integrated theoretical framework drawing on three perspectives: First, Practice Theory examines the everyday routines, interactions, and practices that constitute work. It provides insight into how the integration of AI disrupts established work habits and reshapes professional identities. By focusing on the micro-level activities of software professionals, Practice Theory helps us understand how new practices emerge and evolve in response to technological innovations.
Second, Activity Theory conceptualizes work as an interconnected system comprising subjects, tools (including AI), rules, community, and division of labor. This perspective allows us to analyze the dynamic tensions and interactions that occur when AI is embedded in work processes. It reveals how the collaboration between human and AI agents reconfigures workflows, alters decision-making hierarchies, and transforms the very nature of work tasks. From an institutional perspective, the evolution of software work is marked by a struggle between forces of change and forces of stability. Innovative practitioners—the institutional entrepreneurs—are emerging from the grassroots, leveraging the latest AI technologies to forge new norms and practices in software development. Conversely, more conservative appearance actors, who share many concerns, rely on long-established practices, view these changes as destabilizing, and they work actively to maintain the existing institutional order. This push and pull between change agents and defenders of the status quo creates a vibrant conflict that lies at the heart of the evolving nature of software work.
By integrating these perspectives, our framework provides a rather comprehensive, systemic understanding of the ongoing transformation in software work through different lenses. Every theory has its unique angle that reveals different aspects of change and helps illustrating how AI-driven automation and augmentation are reshaping tasks and roles and to reflect if the underlying rationale and organization of work itself is altering.
Discussion
The proposed framework is conceptual, without direct empirical validation. This tentative conceptual framework offers several implications for future research and practice. Theoretically it opens avenues to critically explore the roles of AI as both a mere tool and an autonomous collaborator. It also invites empirical investigations into how rapid AI advancements affect daily work routines, professional identities, and collaborative practices among software professionals. It also highlights the importance of developing organizational strategies that support sustainable, human-centric work practices—strategies that balance efficiency gains with the need for continuous learning, ethical oversight, and adaptive leadership. Moreover, our framework offers a roadmap for understanding the interplay between technological innovation and organizational change. Addressing AI integration challenges requires a holistic approach that considers both individual worker experiences and institutional dynamics. This approach is crucial for fostering a resilient future for software work, where human creativity and agency are enhanced by AI capabilities.
Conclusions
As AI continues to evolve and transform software development, our integrated framework serves as a robust tool for understanding the complex, systemic changes underway. It underscores that work is a dynamic, multifaceted phenomenon, continually reshaped through the interplay of human practices, technological innovation, and institutional forces. This systemic understanding is critical for guiding future research and for designing practical interventions that ensure work in the AI era remains sustainable, ethical, and centered on human well-being.