The study aims to the impact of social media exposure and behavioural antecedents on fake news sharing behaviour among social media users in Abu Dhabi, UAE using the Theories of Planned Behaviour (TPB) and Reasoned Action (TRA) as the theoretical foundation. The quantitative research methodology using Partial Least Squares – Structural Equation Modelling (PLS-SEM) was employed. The research targeted responses from social media users in Abu Dhabi Emirate. The study found that the background factors of individual, social and knowledge experiences define social media platform exposure of social media users in Abu Dhabi. Exposure to social media has a significant effect on social media users' attitudes and perceived norms, but not perceived behavioural control. Nonetheless, attitudes and norms have no significant effect on behavioural intention to spread fake news. Only perceived behavioural control has an impact on fake news behavioural intention and actual fake news behaviour. The model fit was satisfactory with GoF index of 0.4951, and the SRMR, d_ULS, d-G, Chi-Square, NFI fit indices were all within the acceptable range. Thus, the data fit the theory adequately. The study contributes to the existing body of knowledge by establishing that fake news behavioural intention has a significant influence on social media users' actual fake news behaviour.