Day 2 of the Strategic Competition and Russia resident course, taking place in Garmisch-Partenkirchen, Germany, from Jan. 26 to Feb. 20, 2026, put AI to work as a thinking partner in applied strategy development.
Rather than debating AI in the abstract, participants used commercial large-language models (LLM) directly to support one of the hardest parts of strategy: problem framing. The focus was on how AI can help surface assumptions, expose blind spots, and sharpen problem statements – if it is used deliberately and critically, an essential consideration for security professionals seeking to define problems correctly before proposing action.
“Problem framing is where strategy often succeeds or fails,” said Sae Schatz, Ph.D., Marshall Center team member. “AI, such as commercial large-language models, can help you ask better questions and refine your problem statements. Of course, it’s important to actively use the tools; don’t simply outsource your thinking.
"One effective approach is to use an LLM as a Socratic coach, prompting it to question your assumptions and help you sharpen your own reasoning.”
The day concluded with a hands-on practicum, where participants applied these approaches to real national and institutional challenges, reinforcing that effective strategy starts not with answers but with better questions.