In today’s digital world, fake news presents a serious challenge to social cohesiveness, trust among individuals, and the functioning of democracy. Overcoming this issue demands novel solutions that make efficient use of machine learning (ML) for identifying and battle disinformation. This paper examines the important problem of fake news, including its meaning, its consequences, and the significance of social media in its rapid spread. The study emphasises the complexities of identifying fake news, focusing on sophisticated techniques like AI-generated content and the widespread dissemination of misinformation. In order to tackle such obstacles, the paper uses ML approaches, particularly the LIAR dataset and the XGBoost model, to create a reliable fake news detection system. The outcomes show that these methods have been successful at successfully recognising false information, highlighting ML’s prospective in reducing misinformation. This research contributes to the broader discourse on media literacy and the need for reliable information in the digital age.