A Novel Dataset for Identifying AI Generated Bengali Texts

Md. Siam Ansary · 2025

Ever since artificial intelligence has gained popularity, it is being used and applied to many day to day lives which was quite unimaginable previously. Currently, AI tools are being used to write many texts and documents. Even for preparing academic as well as business reports such platforms are being utilized. AI being able to produce texts and reports is tremendous but it is also important to be able to identify which text is written by humans and which one has been generated using AI platforms. In this research, we have tried to address this issue for one of the most spoken languages on earth which is Bengali. We have created a dataset for the task with texts from various domains and fifty percent of it is AI generated and the rest are human written. All research materials of this study will be made publicly available so that research community can utilize them.

Read the paper · More papers on PaperTik