A total of 3,400 people lost their lives and 14,670 were injured in fires in the USA in 2017, a 9.6 percent increase over the 2007 casualty rate. Floods meanwhile killed more than 100 Americans last year, a number that has also been increasing. To counter this terrible toll, artificial intelligence researchers are developing systems to improve disaster prediction accuracy and provide timely evacuation guidance for panicked people during emergencies. Providing real-time evacuation strategies is critical, as research shows that in emergencies many people tend to wait for instructions when they should already be proceeding to an exit. Simulation systems can play a valuable role in identifying and testing evacuation plans to enable individuals to promptly leave a dangerous area via the safest and fastest route.