This session will explore the considerations and best practices for developing AI tools that are useful, safe, and transparent for actuarial work. While large language models (LLMs) offer exciting possibilities, it is crucial to understand when and how to use them effectively. The session will cover key concepts of LLMs and case studies on developing AI tools for retrieval-augmented generation and summary creation of actuarial documents. Attendees will learn about the importance of designing workflows that nudge users towards safe and effective processes. By the end of the session, participants will gain insights into the responsible use of LLMs in actuarial practice and learn how to leverage their potential to support professional growth.