Smart, But Not Psychic: How AI Accesses the Real World
AI can write you a sonnet but can't tell you the weather - at least not without help. This session shows how language models work around their biggest limitation: they only know what they learned during training. Tools give them access to live information, and MCP is the protocol making it all work smoothly. We'll demo real simple examples. No coding experience needed, just curiosity about what's actually happening when you chat with AI.
Content
This course explores how large language models (LLMs) such as ChatGPT can generate fluent text, code, and explanations, while still facing a key limitation: they do not have direct access to the real world or up-to-date information unless they are connected to external systems. The session explains in a clear and accessible way how these models work, why they cannot “know” things like today’s weather or breaking news on their own, and how the use of external tools helps overcome this limitation. Participants will be introduced to the concept of real-time information access through APIs and external services, as well as the role of MCP (Model Context Protocol) as a standardized way to connect models with tools and data sources. The session includes simple live demonstrations, focusing on what actually happens behind the scenes when interacting with modern AI systems.
Target Audience
University lecturers, students, researchers, technology professionals, and anyone curious about Artificial Intelligence who wants to understand how language models connect to external tools and real-world information. No coding experience is required—only curiosity about what is really happening when we chat with AI.
Motivation
As AI becomes increasingly embedded in everyday life, it is essential to understand not only what language models can generate, but also what they cannot know. Many users assume that chatbots automatically have access to the internet or real-time data, when in reality this depends on specific tool integrations. This course aims to demystify that process and explain how AI systems can safely and effectively access real-world information through external tools. By understanding these mechanisms, participants will gain a more critical and informed perspective on modern AI and the emerging technologies that are shaping the future of intelligent applications.