AI-generated Essays: Characteristics and Implications on Automated Scoring and Academic Integrity

TL;DR

This paper analyzes AI-generated essays' characteristics and their impact on automated scoring and academic integrity, highlighting limitations of current systems and potential detection strategies.

Contribution

It provides a large-scale empirical benchmark of AI-generated essays and evaluates the effectiveness of detection methods across different language models.

Findings

Existing automated scoring systems struggle with AI-generated essays.

Detectors trained on one model's essays can often identify others.

AI detection remains feasible despite increasing model variety.

Abstract

The rapid advancement of large language models (LLMs) has enabled the generation of coherent essays, making AI-assisted writing increasingly common in educational and professional settings. Using large-scale empirical data, we examine and benchmark the characteristics and quality of essays generated by popular LLMs and discuss their implications for two key components of writing assessments: automated scoring and academic integrity. Our findings highlight limitations in existing automated scoring systems, such as e-rater, when applied to essays generated or heavily influenced by AI, and identify areas for improvement, including the development of new features to capture deeper thinking and recalibrating feature weights. Despite growing concerns that the increasing variety of LLMs may undermine the feasibility of detecting AI-generated essays, our results show that detectors trained on essays generated from one model can often identify texts from others with high accuracy, suggesting that effective detection could remain manageable in practice.