Ever been in a conversation with someone claiming to know what you've just told them, but you're not entirely sure if they're bluffing? 🤔 Turns out, Large Language Models can bluff too! In this shot, we're diving deep into the world of LLM honesty. We'll explore how to determine if an LLM truly knows the answer to a question or if it's simply bluffing its way through. Using a unique dataset of personal questions about a synthetic persona, we'll put two different Gemini models to the test. We'll also visualize our findings with a cool Sankey diagram! Join us as we uncover the surprising truth behind LLM "knowledge" and learn how to call their bluff! This shot was filmed in front of a live audience 👥👥, as part of Bon Voyage Shot 🇮🇱🇺🇸 (https://www.oneshotlearning.io/bon-voyage-shot). We'd like to thank AWS Floor28 for their warm hospitality.