From Data to Decisions with LLMs: AI-Powered Statistical Leadership & Collaboration

In today's data-driven environment, the role of statistical leaders is expanding beyond technical expertise to effective communication, cross-functional collaboration, and strategic decision-making across academia, industry, and government. To meet these demands, artificial intelligence—especially large language model (LLM) tools such as ChatGPT and Copilot—has become a practical aid. These assistants enhance workflows, increase productivity, and foster innovation in statistical practice, with particular relevance to biopharmaceutical and regulatory settings. This short course offers a concise, software-agnostic exploration of how LLM assistants can strengthen statistical leadership and collaboration. Participants will learn core capabilities and practical applications: effective prompting; improving communication with non-statistical stakeholders; supporting decision-making; and integrating assistants into R and SAS workflows. We also address ethical and scientific considerations—mitigating biases, ensuring reliability and repeatability, validating outputs, and maintaining transparent documentation and provenance for AI-assisted analyses. Through multiple case studies and interactive exercises, attendees will gain practical skills and reusable artifacts (checklists, patterns, and templates) for enhancing stakeholder engagement, initiating adaptive designs, conducting simulations, and making strategic decisions. By enabling statistical leaders to leverage LLM tools responsibly, the course equips participants to tackle complex challenges, streamline processes, strengthen collaboration, and make informed decisions with greater confidence.