Computer technologies have increased manifold, causing crime activities online to become more prolific thus presenting a complex investigation procedure that would otherwise be hard to resolve in the aid of the traditional digital forensics. The paper explains the application of artificial intelligence (AI) to enhance the digital forensic process regarding detection levels, speed of analysis, and outcomes of an investigation in several crimes associated with cyber-related crimes. The study compares the performance of multiple AI models, such as convolutional neural networks (CNNs), support vector machines (SVMs), random forest classifiers, natural language processors (NLP) models, and artificial neural networks (ANNs), based on a structured dataset and comparative analysis of their results. Findings demonstrate that structured evidence, e.g., executable files and encrypted information, allows AI to detect more accurately, whereas unstructured evidence, e.g., network traffic, increases the time of processing and reduces performance. The effectiveness and success rate of investigations of CNN and SVM models were shown to be highly efficient and more successful than simple models. It is discussed that adaptive, ethically appropriate AI systems are required, capable of managing various types of digital evidence. Results show that AI can be used to revolutionize forensic operations through automation, enhancing the detection of threats, and assisting in investigative decision-making. Nevertheless, there are still constraints in examining socially engineered attacks and the high volume of unstructured information. Overall, the present paper has demonstrated that the use of AI-based solutions is significant in the establishment of modern digital forensics.