This investigation intended to predict the probability of bankruptcy in the Iranian and Iraqi stock exchanges, utilizing neural networks and machine learning techniques during the COVID-19 crisis. The investigation considered two separate time periods: the pre Covid-19 period, which extends from 2003 to 2019, and the post Covid-19 financial crisis period, which extends from 2020 to 2023. The primary concern was creating a model that would be employed in the forecasting of bankruptcy risk in Iranian and Iraqi banks. Different machine learning models, including decision trees, XGBoost, and random forests, were employed in this study. The key variables in the study of how to predict the risk of bank failure were found to be LTICA, ILQ, LRG, and CLRE. These variables were used to predict the risk of bank failure. The performance of neural networks and support vector machines (SVMs) increased during the pre- and post Covid-19 eras, this was attributed to the increased prevalence of crises. It was confirmed that the forecast accuracy of the SVM model increased by 0.9321 during the period of Covid-19, while it was still 0.9125 before the crisis. Other improvements were apparent in the indices of sensitivity and specificity from all three classes during the crisis. The results suggested that it's very important to utilize multiple approaches in predicting the risk of bank failure. This is because combining different machine learning methods and neural networks increases the effectiveness of recognizing the causes of company bankruptcy. This also demonstrates the significant influence of macroeconomic variables, such as GDP and interest rates, in the models associated with predicting the probability of a bank's failure. The investigation focuses on the value of government assistance during crises that were augmented by the enhanced risk management prowess of Banks during the Cobra era (Covid-19). Practical suggestions derived from this research include the creation of an early warning system, enhancement of risk management methods, and regular training for the employees of banks and their supervisors.