Novice software students review and adjust AI outputs beyond correctness
Beyond Correctness: How Novice Software Engineering Students Exercise Oversight of Generative AI
Software EngineeringHuman-Computer Interaction
Summary
Software engineering students working with AI tools don’t just check for mistakes; they also judge how well AI-generated solutions fit their project needs. The authors studied how first-year students recorded AI interactions and found that students often focused on whether the AI’s answers were suitable rather than just correct. Students frequently changed AI outputs or rephrased their questions to get better results. This shows that early learners exercise judgment and take active roles in supervising AI as part of their work.
What this means in practice
- •For software engineering educators: Design course materials that help students develop skills in judging and refining AI-generated code beyond checking for correctness.
- •For software development teams: Incorporate structured oversight practices where developers review AI outputs for suitability within project contexts, not just correctness.
Authors
Ali Shakiba, Thomas Chaffey
Abstract
Generative AI (GenAI) is shifting software engineering students' work from producing artefacts towards evaluating and supervising AI-generated outputs. However, little is known about what characteristics students identify in outputs when reporting interventions, or how they exercise oversight during authentic software engineering work. We analysed 962 student-AI interaction turns documented by 145 first-year software engineering students through structured AI Journals, including 172 student-declared interventions. Using a multi-dimensional coding framework, we characterised the output characteristics identified in these reported interventions, students' reasons for intervening, the actions they took, and how those actions were enacted. Fitness problems were identified in 74.4% of interventions, compared with 16.9% involving correctness problems, indicating that oversight more often concerned whether plausible AI outputs were suitable for the engineering task than whether they contained explicit errors. Among classifiable interventions, students most frequently modified generated outputs (51.0%) or refined their prompts (28.2%), with both response types commonly associated with engineering fit. Within-student analysis further showed that outputs perceived as containing minor flaws were associated with lower evaluation confidence than outputs perceived as perfect. Taken together, these findings characterise novice AI oversight as a process that extends beyond error detection to encompass both judgement of fitness for purpose and active intervention in AI-assisted engineering work.