Abstract: The internet and social media have really changed the way we make and share information. We can talk to people faster. Reach people all around the world. There is a bad side to this. It is easy for fake news to spread. Fake news is when someone makes up a story or lies, on purpose. Says it is real news. They do this to trick people. When fake news spreads quickly it can cause a lot of problems. These problems include people not agreeing on things not trusting each other being upset and angry and fake news can even affect how our countries are run. Fake news is a deal because it can really hurt people and communities. The spread of news can lead to people not trusting the news at all and it can also lead to people fighting with each other. Fake news is a problem that we need to worry about because it can affect our societies and our democratic systems. The old way of checking facts by hand is good.Fake news detection systems that use intelligence look at news to see if it is real or not. They use machine learning and other things like learning and natural language processing to do this. These systems look at a lot of news stories that are labeled as real or fake to learn what makes them different.They learned from ways of doing things like Logistic Regression and other methods to figure out what makes news fake. They looked at things like how the newss written and what words are used to decide if it is real or not. We have made a lot of progress.This is a problem because we need to be able to trust intelligence systems and understand how they work.Nevertheless, AI-based fake news detection systems offer substantial societal benefits. They support content moderation, assist journalists in verification, and help protect public discourse. Future research aims to improve model robustness, enhance explainability, and strengthen collaboration between human experts and AI systems.

Keywords: Natural Language Processing -Fake News Detection-Logistic Regression-Content Moderation-Fact-Checking- Explainable AI

Downloads: | DOI: 10.17148/IARJSET.2026.13238