Watch this invited talk from the 27th International Symposium on Practical Aspects of Declarative Languages (PADL'25) where Umut A. Acar from Carnegie Mellon University challenges the traditional trade-off between safety and performance in programming languages. Discover how parallel functional programming languages can deliver impressive performance while maintaining safety, and learn how programming languages can automatically optimize their performance without sacrificing safety. The presentation explores a methodology that combines algorithmic complexity theory, programming language theory, and empirical computer science to achieve both safety and performance goals simultaneously. This 57-minute talk, sponsored by ACM SIGPLAN, addresses the increasing societal and economic importance of software safety while demonstrating practical approaches to overcome the historical dichotomy between safe and performant code.