Have you ever wondered what really happens when artificial intelligence (AI) writes text, generates images, or answers your questions?
If you’ve ever been curious about how ChatGPT “knows” what to say, how realistic images are created from text, or why AI sometimes gets things right and other times completely misses the point or starts to “hallucinate” — this workshop is your chance to understand it all through simple explanations and concrete examples. This session reveals how AI actually “thinks,” demystifying the core principles behind the technology that’s transforming the world.
Together, we’ll break down how large language models are trained on massive amounts of data, how they learn to recognize patterns and generate responses, and why — despite all that — they still don’t understand reality the way humans do. Through interactive demonstrations, you’ll see what happens “behind the scenes” as AI predicts the next word, creates an image, or generates an idea — and why it can sometimes feel like a superpower, while other times it goes completely off track.
We’ll discuss the current limits of the technology, the most common mistakes AI makes, and how you — as a user — can recognize and avoid potential pitfalls. You’ll learn practical ways to use GenAI as a tool that saves time, boosts creativity, and supports your work, learning, and research — all while keeping a clear understanding of where its “intelligence” ends and yours begins.
Because after this workshop, you’ll start using AI as a powerful ally. This isn’t a lecture for programmers — it’s for anyone who wants to understand, analyze, and use AI responsibly, wisely, and with a healthy dose of curiosity.