The proliferation of fake bots on social media platforms poses a significant threat to the integrity of online discourse. These automated accounts often engage in malicious activities such as spreading misinformation, manipulating public opinion, and amplifying divisive content. This paper presents an explainable AI framework that leverages DistilBERT, a lightweight and efficient transformer model, for the task of fake bot detection. Our approach goes beyond simple prediction; we utilize attention visualization and feature importance analysis to interpret the model’s decision-making process. By highlighting the key behavioral and linguistic patterns driving the model’s predictions, we aim to increase transparency and trust in AI-based bot detection systems. Experimental results demonstrate the effectiveness of our framework in identifying salient features, such as repetitive posting patterns, unusual activity times, and the use of coordinated language, that contribute to the model’s ability to discern between real users and fake bots. This work offers valuable insights into the strengths and limitations of the DistilBERT model for fake bot detection, paving the way for more robust and interpretable AI solutions to combat the spread of malicious automated accounts on social media.