Florian Dirisamer's presentation at the NAFEMS conference, titled "Revolutionizing Computer-Aided Engineering: The Use of AI in Optimizing FEM and CFD Simulations and Innovatively Handling Inverse Problems and Black-Box Components with Examples from Simulations of Water Injection Technology in Plastic Injection Molding," explored the application of Artificial Intelligence (AI) in enhancing the efficiency of engineering simulations. The focus of Dirisamer's talk was on the utilization of AI in process simulations, particularly in the context of water injection technology in plastic injection molding. He demonstrated how AI significantly speeds up the process by replacing parameter studies and simplifying the resolution of optimization problems through the inversion of questions. This approach was shown to be not only faster but also more reliable in predicting outcomes. The presentation detailed the comparison between the state-of-the-art approach and the AI approach in handling process and geometry parameters, as well as in analyzing simulation and measurement results. A critical aspect of the AI approach was the training process, where a significant emphasis was placed on the accuracy of predictions, achieving an impressive 90% accuracy in training and 10% in evaluation. Dirisamer highlighted the goal of the AI approach in identifying the parameter window to ensure the required wall thickness in plastic parts. This involved a detailed process of determining the coefficient of determination to compare the original values from simulation results with the AI-predicted values, optimizing the AI structure through iterative loops. The inversion of the problem statement was another key topic, where the AI's ability to predict process parameters for optimized results was showcased. This approach saves time and resources, illustrating the practical benefits of AI in engineering simulations. Dirisamer concluded with a comparison between simulation results and AI predictions, underscoring the efficacy of AI in closely matching the simulated outcomes.