Normative boundaries of AI in scientific work: Evidence from PhD researchers

Summary

Latent class analysis of 3,785 PhD students from Nature's Graduate Survey 2025 recovers four attitudinal profiles toward AI across five research tasks. The dominant profile is an explicit 'division of labour': AI accepted for tracking and summarising literature, resisted for writing, data analysis and experimental design, i.e. exactly the tasks tied to intellectual contribution, authorship and responsibility. Gives empirical shape to the human-AI division-of-labour question rather than asserting it, and draws out implications for governance, doctoral training, disclosure and research evaluation.

Contribution

Queued — not yet read.

Method

Queued — not yet read.

Relevance to RISE

Surfaced by the 2026-09-07 literature sweep as a high-priority candidate for the RISE knowledge base (angle: human–AI research collaboration). Verified against: Opened https://arxiv.org/abs/2608.25678, read full abstract, 2 authors, 2026-08-26, econ.GN. Dedupe: no 'normative' match.

Critique / open questions

Queued — not yet assessed.

Key quotes

Artificial intelligence (AI) is increasingly embedded in scientific work, but researchers may not evaluate its use uniformly across research tasks. (abstract)

This study examines task-specific attitudes towards AI among an international, self-selected sample of 3,785 PhD students in STEM and medical and health sciences who participated in Nature's Graduate Survey 2025. (abstract)