With the advent of technology, everything has become digitized, including newspapers and magazines. Currently, information is accessible in an easy, and fast manner. However, some content creators exploit this opportunity negatively by using unethical methods to attract users' attention with the objective of increasing their ads' income rather than providing trusted information. To address this clickbait phenomenon, we propose various approaches based on natural language processing and deep learning models to detect this type of content in Arabic. The results showed that the fine-tuned BERT model combined with an attached neural network layer or with a self-attention network provides similar performance in terms of accuracy: 91.86\% and 91\% respectively compared to RoBERTa, Word2vec, and TF-IDF using CNN, LSTM, and Neural network. The collected dataset is sourced from multiple Arabic websites.