Ozancan Ozdemir

Ozancan Ozdemir

PhD Candidate at Bernoulli Institute for for Mathematics, Computer Science and Artificial Intelligence. Machine Learning and Deep Learning Enthisuasist.

From Perceptions to Evidence: Detecting AI-Generated Content in Turkish News Media with a Fine-Tuned BERT Classifier

Published in arXiv, 2026

Recommended citation: Ozdemir, O. (2026). From Perceptions to Evidence: Detecting AI-Generated Content in Turkish News Media with a Fine-Tuned BERT Classifier. arXiv:2602.13504. https://doi.org/10.48550/arXiv.2602.13504 https://arxiv.org/abs/2602.13504

Overview

This preprint studies the detection of AI-generated content in Turkish news media using a fine-tuned Turkish BERT classifier. It moves beyond self-reported perceptions of AI use in journalism and offers an empirical, data-driven approach to measuring AI-rewritten content across Turkish news outlets.

Read the preprint

PDF

Citation

Ozdemir, O. (2026). From Perceptions to Evidence: Detecting AI-Generated Content in Turkish News Media with a Fine-Tuned BERT Classifier. arXiv:2602.13504. https://doi.org/10.48550/arXiv.2602.13504