RogueGPT: A controlled stimulus generator for AI news authenticity research. (arXiv:2601.21963 and arXiv:2601.22871)

The aim of this project is to generate fake news in the Azerbaijani language using LSTM Recurrent Neural Networks. LSTM Recurrent Neural Networks are powerful Deep Learning models which are used for learning sequenced data. Here a LSTM model was trained on 65 thousand samples, and it should be able to generate text.

End-to-end Fake News Detection & Generation system using GPT-2 for headline generation and DistilBERT for real/fake classification. Features data preprocessing, model training, FastAPI backend, Streamlit UI, and cloud deployment. Evaluated using Accuracy, Precision, Recall, F1, Perplexity, Distinct-n, and Self-BLEU.

Exploring the Dual Role of AI: RNNs and BERT for Fake News Detection and LLMs for Fake News Generation

A simple Python script that generates funny random “breaking news” using names, actions, and places — with text-to-speech for extra chaos. My first Python learning project.