Published Paper Details:

The rapid expansion of digital media has significantly increased the demand for timely, accurate, and structured news content. Traditional newsroom workflows rely heavily on manual writing, editing, and verification processes, which are time-consuming and prone to human fatigue and bias. To address these limitations, this paper presents an Auto Newsroom Agent System, an intelligent multi-agent framework designed to automate the end-to-end news generation pipeline. The proposed system employs a Multi-Agent System (MAS) architecture consisting of four specialized agents: a Writer Agent based on the GPT-2 transformer model for coherent article generation, a Fact-Checker Agent utilizing a RoBERTa-based classifier for identifying potential misinformation or sensitive content, a Headline Generator Agent for concise and context-aware headline creation, and a central Editor Agent that orchestrates the overall workflow. The system accepts user-provided topics or RSS feed summaries, generates structured news articles, verifies content integrity, and stores finalized outputs in a persistent SQLite database through a Flask-based web interface. Experimental evaluation demonstrates that the system can generate complete news articles within seconds while maintaining structural consistency and basic content verification. The modular design ensures scalability and extensibility, enabling future integration of advanced fact-checking and multimedia generation components. The Auto Newsroom Agent System highlights the potential of AI-driven automation in digital journalism, offering an efficient and cost-effective solution for modern news production.