AI Literacy Series: A Practical Guide to Modern AI Terms

Understanding modern AI increasingly requires more than recognizing the word model. Meetings, vendor presentations, articles and podcasts now routinely use terms such as tokens, RAG, agents, MCP, guardrails, benchmarks and inference cost. Knowing how these ideas relate makes it easier to judge what a system actually does, what evidence supports its claims and where human oversight is still required. 

In this course, Jason Dean builds that vocabulary from the ground up. He explains 57 widely used AI terms in direct, non-technical language, using practical examples from invoices, contracts, finance and workplace software. The course begins with models, tokens and training; moves through prompts, context, retrieval and grounding; then examines APIs, tools, agents, evaluation, security, openness, latency and cost. Clear diagrams and animations show what happens at each stage and how the terms connect. 

The course is designed for professionals who use, evaluate, procure, manage or write about AI. By the end, you should be able to follow an AI meeting, vendor presentation, article, podcast or news report with confidence. You should also be better prepared to ask precise questions about what a system can do, how it was evaluated, what controls surround it and what it will cost to operate. 

By the end of the course, learners should be able to: 

- follow AI meetings, vendor presentations, articles and news coverage without stopping to translate every term;

- distinguish a model from the application and controls around it;

- explain how training, inference, prompting, retrieval, tools and agents relate;

- interpret common claims about benchmarks, security, openness, latency and cost; and

- ask more precise questions when evaluating an AI product or proposal.