Reviewing the performance of AI detection tools in differentiating between AI-generated and human-written texts: A literature and integrative hybrid review

In short

Pulling together 17 articles from 2023 on how well AI-writing detectors told machine text from human text, the review reports the studies found every tool inconsistent and recommends adding human reviewers.

What it looked at

A systematic-style review of 17 articles published in 2023, guided by six research questions and found through eleven databases, two search engines and one academic networking site. The reviewed articles came from twelve countries, and every one used ChatGPT, in its GPT-3.5 or GPT-4 version, to produce machine text, sometimes alongside other generators.

What it means for teaching

Human review alongside any detector result is the review's recommendation, since the reviewed studies found every tool they tested inconsistent.

Limits

Covers only a short publication window in 2023 and only the tools and chatbot versions those articles tested. How comparable the underlying test texts were is not described.

Cited 85 times by other papers, as counted by OpenAlex in November 2026. A count measures attention, not quality.