This study aims to investigate the role of applying artificial intelligence in limiting cybercrime in the 13 Jordanian commercial banks listed in the Amman Financial Market, from the point of view of internal auditors and IT. For this purpose, a questionnaire was designed and distributed to the study sample of (849). The number of respondents to the questionnaire reached (230). The data of the questionnaire were analyzed, and their hypotheses were tested using the statistical program (SPSS) through tests of the arithmetic mean, standard deviation, linear, multiple regression, and the T-test for two independent samples. The study found a statistically significant effect for artificial intelligence with its dimensions (Expert Systems, Artificial Neural Network, Genetic Algorithm, Fuzzy logic) in limiting cybercrime in Jordanian commercial banks. It was also found that there were statistically significant differences in both genetic algorithm and cybercrime attributable to the job variable in favor of IT department employees.

Social Engineering (SE) Attacks against information systems continue to pose a potentially devastating impact. Security information systems are becoming increasingly significant as the number of SE incidents rapidly increased and became more aggressive than before. The World Wide Web (WWW) has evolved for information exchange and knowledge-sharing. It enables the sharing of information in a timely, effective, and transparent manner. Identity theft and identity misuse are two sides of cybercrime in which hackers and fraudulent users collect sensitive information from current legal users in order to perform fraud or deceit for financial gain. Malicious links are used as phishing methods, in which malicious links are planted beneath legitimate-looking links. As the number of web pages grows, the number of malicious web pages and the attacks of such become more complex. In this paper, we provide a method for identifying malicious web pages using a crawling and classification approach that helps to support the automatic discovery of the malicious links. The proposed approach can successfully complete the crawling session even if the page requires partial page refreshment and authentication credentials. The evaluation of the proposed approach shows a higher accuracy compared to an existing approach with an overall accuracy of 72% in three custom applications. Moreover, the proposed approach will calculate the significance and the impact severances of each link on the website and it better differentiates malicious web pages and normal links. The results of the proposed approach will also help in providing a set of recommendations which can increase the awareness level of the end-users, website administrators on how to better deal with these types of SE attacks.