Social engineering is the manipulation of people’s virtual belongings and misusing them by unauthorised access. This article looks at aspects for pertinent to value, losses, social practices and then offers a defence mechanism framework. Social engineering can cause considerable damage to an individual or an organisation before they are detected. Consequently, ample defence mechanisms should be in place to prevent such attacks. Information security threats are as much technical as they are social. Social engineering attacks aim to gain unauthorised access to information. A hacker’s cunning manipulation of an individual’s nature of trust to obtain information would grant the hacker unauthorised access to the virtual data of an individual for a predatory stance. Social engineering can take place over various platforms, and social media and social networking are among them. Social platforms, with their distracting, defocussing potential combined with the fact that is can also be used as a means for executing a social media attack. The proposed framework in the paper targets threats in social engineering; it is about tricking people into providing access to information. It addresses the serious losses the attacks can create and the necessity of defences. Psychological motives are the focus of the framework, and a novel method based on AI prevents these threats. This study fist explores the psychological modes and offers a contextual AI-based solution framework which counter the social engineering mechanisms and with their integration into cyber security policies, these can defect deep levels of intrusions. This research benefits society by raising awareness of social engineering risks and enhancing cybersecurity measures to safeguard information.