Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AI Where It Matters: Where, Why, and How Developers Want AI Support in Daily Work

Rudrajit Choudhuri, Carmen Badea, Christian Bird, Jenna Butler, Rob DeLine, Brian Houck

TL;DR

This study addresses the lack of guidance on where and how to deploy AI support in software engineering by applying cognitive appraisal theory to a large mixed-methods dataset (N=860) of Microsoft developers. It identifies four task appraisal drivers—Value, Identity, Accountability, and Demands—and shows how they predict openness to and use of AI, revealing distinct task clusters with different AI needs. The work also maps developers' prioritized Responsible AI principles across task categories and dispositions, highlighting a focus on reliability, privacy, transparency, steerability, and goal maintenance, with more emphasis on fairness and inclusiveness for human-facing work. The findings offer concrete guidance for designing task-aware, augmentation-focused AI tools that preserve developer agency and craft, with implications for practice and future research on observability, guardrails, and adaptive AI design in software engineering.

Abstract

Generative AI is reshaping software work, yet we lack clear guidance on where developers most need and want support, and how to design it responsibly. We report a large-scale, mixed-methods study of N=860 developers that examines where, why, and how they seek or limit AI help, providing the first task-aware, empirically validated mapping from developers' perceptions of their tasks to AI adoption patterns and responsible AI priorities. Using cognitive appraisal theory, we show that task evaluations predict openness to and use of AI, revealing distinct patterns: strong current use and a desire for improvement in core work (e.g., coding, testing); high demand to reduce toil (e.g., documentation, operations); and clear limits for identity- and relationship-centric work (e.g., mentoring). Priorities for responsible AI support vary by context: reliability and security for systems-facing tasks; transparency, alignment, and steerability to maintain control; and fairness and inclusiveness for human-facing work. Our results offer concrete, contextual guidance for delivering AI where it matters to developers and their work.

AI Where It Matters: Where, Why, and How Developers Want AI Support in Daily Work

TL;DR

This study addresses the lack of guidance on where and how to deploy AI support in software engineering by applying cognitive appraisal theory to a large mixed-methods dataset (N=860) of Microsoft developers. It identifies four task appraisal drivers—Value, Identity, Accountability, and Demands—and shows how they predict openness to and use of AI, revealing distinct task clusters with different AI needs. The work also maps developers' prioritized Responsible AI principles across task categories and dispositions, highlighting a focus on reliability, privacy, transparency, steerability, and goal maintenance, with more emphasis on fairness and inclusiveness for human-facing work. The findings offer concrete guidance for designing task-aware, augmentation-focused AI tools that preserve developer agency and craft, with implications for practice and future research on observability, guardrails, and adaptive AI design in software engineering.

Abstract

Generative AI is reshaping software work, yet we lack clear guidance on where developers most need and want support, and how to design it responsibly. We report a large-scale, mixed-methods study of N=860 developers that examines where, why, and how they seek or limit AI help, providing the first task-aware, empirically validated mapping from developers' perceptions of their tasks to AI adoption patterns and responsible AI priorities. Using cognitive appraisal theory, we show that task evaluations predict openness to and use of AI, revealing distinct patterns: strong current use and a desire for improvement in core work (e.g., coding, testing); high demand to reduce toil (e.g., documentation, operations); and clear limits for identity- and relationship-centric work (e.g., mentoring). Priorities for responsible AI support vary by context: reliability and security for systems-facing tasks; transparency, alignment, and steerability to maintain control; and fairness and inclusiveness for human-facing work. Our results offer concrete, contextual guidance for delivering AI where it matters to developers and their work.

Paper Structure

This paper contains 21 sections, 2 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Appraisal Foundations & Hypotheses
  4. Method
  5. Study Design
  6. Data Collection
  7. Data Analysis
  8. Results
  9. RQ1a: How do appraisals shape openness to and use of AI support?
  10. RQ1b: Where and why do developers seek or limit AI support?
  11. Core work (C1)
  12. People & AI-Building (C2)
  13. Ops & Coordination (C3)
  14. RQ2: Which RAI design principles do developers prioritize in AI for SE tasks?
  15. How do priorities vary across task categories?
  16. ...and 6 more sections

Figures (2)

  • Figure 1: Scatter z-score plot showing relative AI-support needed ($z$) versus Usage ($z$) scores for SE tasks. Tasks are grouped into four quadrants representing strategic zones: Build, Improve, Sustain, and De-prioritize.
  • Figure 2: Participants (%) selecting each RAI principle as their top-5 priority for AI support across SE task categories; percentages reflect frequency and do not sum to 100%.