.png){kind=icon}
.png){kind=icon}
Return
Auditing unauthorized training data from AI generated content using information isotopes
.png){kind=icon}
OA
AI
DOI：10.1038/s41467-026-68862-x.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The rapid growth of AI systems has been fueled by large-scale human data, intensifying concerns over the unauthorized use of intellectual property and privacy-sensitive content during model training. Auditing such misuse is particularly challenging since mainstream AI services operate as black boxes, exposing only generated outputs while concealing their training and inference processes. In this work, inspired by chemical isotope tracing, we introduce the concept of information isotopes to trace training data within opaque AI systems. We propose an information-isotope tracing framework that selectively marks target data elements and detects their propagation in model outputs, providing concrete evidence of data utilization under black-box access. Experiments on thirteen AI models across six datasets demonstrate that our method distinguishes training from non-training data with up to 99% accuracy and strong statistical significance (p < 0.01) using approximately 4,000 words of evidence. An open-source tool is released to support practical data rights protection. Here, the authors introduce information isotopes, a novel framework for tracing unauthorized training data in black-box AI systems, offering a practical tool for auditing data provenance and protecting data rights.
Keywords:
Computer science
Ethics
Intellectual-property rights
Science
Humanities and Social Sciences
multidisciplinary
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
15.7
Papers:
9.2W
Citations:
91.2W
View more organizations