When AI Blurs the Boundaries of Contribution: An Empirical Study of Authorship Calibration

Source

Source name: arXiv
Source kind: paper
Authors / organization: Célina Treuillier, Denis Lalanne
Venue: arXiv: cs.HC
Published: 2026-07-16
Collected: 2026-07-17
URL: http://arxiv.org/abs/2607.15006v1
PDF: https://arxiv.org/pdf/2607.15006v1
DOI: UNKNOWN

Plain-Language Brief

This research investigates how people's understanding of their own authorship changes when using AI, particularly Generative AI. The study uses the CoAuthor dataset and finds that frequent users tend to overestimate their contribution compared to less frequent users. This misjudgment can have implications for learning outcomes.

What the Document Says

The paper discusses the concept of authorship calibration, which refers to a user's awareness of their actual authorship when interacting with AI.
The study uses the CoAuthor dataset and empirically examines authorship calibration across users based on their frequency of AI use.
The results show that frequent AI users tend to misjudge their authorship, while less frequent users exhibit more accurate authorship calibration.

Key Concepts Mentioned

Artificial Intelligence
Generative AI
Authorship Calibration
CoAuthor dataset

Claims Made by the Document

AI can affect a user's perception of their own authorship.
Misjudgment of authorship can impact learning outcomes.

Evidence Used

Findings reveal high variability in authorship calibration across users based on their frequency of AI use.

Methods / Approach

Empirical study using the CoAuthor dataset

Limitations Stated by the Authors

Not stated in source

Limitations Noted by the Agent

Agent interpretation — not from the source.

The study is limited to the findings from the CoAuthor dataset, and it's unclear if these results would generalize to other datasets or types of AI.

Open Questions

How does authorship calibration vary with different types of Generative AI?
What are the long-term effects of misjudging authorship on learning outcomes?

Notes for Future Q&A

May include interpretation. Labeled where it does.

The paper focuses on Generative AI, but it's unclear if other types of AI would have similar effects on authorship calibration.

Raw Abstract / Excerpt

The broad adoption of Artificial Intelligence (AI), especially Generative AI, raises pressing questions about how users interact with these systems to produce new content. In this paper, we introduce the concept of authorship calibration, defined as users awareness of their actual authorship when interacting with AI. Using the CoAuthor dataset, we empirically examine how authorship calibration varies across users and how it relates to their frequency of AI use. Our results reveal high variability: users relying heavily on AI tend to misjudge their authorship, whereas those using AI less frequently exhibit more accurate authorship calibration. These findings suggest that AI can obscure users perception of their own authorship. In learning contexts, miscalibration can affect metacognitive monitoring and learning strategies, ultimately impacting learning outcomes. Fostering authorship calibration then appears essential for promoting responsible and educationally meaningful AI integration.