[Submitted on 3 Jun 2026 (v1), last revised 19 Oct 2026 (this version, v2)]
Abstract:Although it is generally agreed that AI-generated text poses a broad societal risk, there is no common understanding in the AI-generated text detection literature on what constitutes harmful use. Rather, existing datasets and approaches often define their own criteria and make their own assumptions, sometimes implicitly, and often only loosely related to real-world needs and applications. To address this gap, we here systematically define various notions of AI-generated text and their characteristics. To study these, we collect AITDNA – a new benchmark of human-machine co-constructed texts that is annotated with detailed genesis information, such as the entire edit and AI-interaction history. We benchmark various machine-generated text detectors and find that they often only perform well for specific notions but not as broad detectors. We release code and data publicly.
Submission history
From: Marina Sakharova [view email]
[v1] Wed, 3 Jun 2026 14:05:14 UTC (663 KB)
[v2] Mon, 19 Oct 2026 14:14:54 UTC (669 KB)
