What question did this study set out to answer?

The study aims to compare text readability and writing styles between human-written and AI-generated academic abstracts.

April 10, 2026PLoS ONEOpen Access

Comparative analysis of text readability and writing styles in AI-generated vs. Human-written academic abstracts

View Full Paper

Ask AI

Bookmark

Authors

This comparative analysis assesses readability in AI-generated and human-written abstracts, suggesting AI's limitations in complexity.

The study aims to compare text readability and writing styles between human-written and AI-generated academic abstracts.
Analyzed 150 human-written abstracts from linguistics and computer science journals.
Generated 150 corresponding AI-generated abstracts from the same corpus.
Used the Readability Scoring System for metric evaluation.
Conducted expert reviews to assess quality of AI-generated abstracts.
Applied SPSS 27 for non-parametric statistical analyses.
AI-generated abstracts showed significantly lower readability scores across eight metrics.
Five metrics differed in readability for linguistics, while eight differed in computer science.
Interdisciplinary comparisons indicated non-significant differences across nine metrics, suggesting AI's mimicry of natural writing.
AI-generated abstracts face challenges in achieving lexical diversity.

Zou et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark