AI: How It Works and How to Work With It
Wednesday, February 11 | 9am - 11am | In-Person
Friday, February 13 | 9am - 11am | In-Person
When you ask an AI like ChatGPT a question, what is it actually doing? According to recent surveys, 45% of people think it's looking up answers in a database. 21% think it's following a script of prewritten responses. Only 28% understand that it's predicting what words should come next based on patterns it learned.
This misconception matters. If you don't understand how AI works, you can't work with it effectively. You'll write prompts that fail, trust outputs you shouldn't, and miss capabilities that could transform your work.
This workshop strips away the mystery. You'll learn:
How AI Actually Works
- What "generative AI" and "large language models" really mean
- The key concepts: training, tokens, probability, and generation
- Why the best practices are just that and how to internalize them
The Key AI Tools
- ChatGPT, Claude, Gemini, and when to use each
- Perplexity for search, NotebookLM for synthesis
Capabilities and Limitations
- What AI can do: understanding, generation, transformation, retrieval, analysis, discussion, reasoning
- What AI can't do (yet): context window limits, training cutoffs, the "aiming to please" problem
- Agentic AI: deep research, browser agents, code execution
Best Practices
- Know your tools, give context, assign roles
- Ask, then task—the key to better outputs
- Step-by-step reasoning and the power of examples
- Personalization: custom instructions, memory, projects, and context engineering
No prior AI experience is required—just curiosity and a willingness to learn. Bring your laptop to participate in hands-on activities.
Collaboration with CELT and CAIDS.
