Lesson 1 of 5Beginner 9 min

6 AI concepts everyone talks about, explained simply

Key takeaways

  • LLMs are the engines behind chat assistants like ChatGPT, Claude and Gemini
  • RAG lets an AI look things up in your own documents before it answers
  • Agents take actions, and MCP is the standard that connects them to other tools
  • System prompts are hidden instructions that shape how an assistant behaves

If words like LLM, RAG, agent or MCP make you nod along without really knowing what they mean, start with this one. In under nine minutes, IBM's Jeff Crume walks through six concepts that sit behind almost every AI product you'll use, and he explains each one with a plain human analogy instead of math.

What it covers

The video starts with large language models (LLMs), the engines behind ChatGPT, Claude and Gemini, and what they actually do when they "answer" you. From there it moves to RAG (retrieval-augmented generation), the trick that lets an AI look things up in your own documents before it replies. Then come AI agents, which don't just talk but take actions, and MCP, the standard that lets agents plug into other tools. It finishes with system prompts (the hidden instructions that shape how an assistant behaves) and how models are trained in the first place.

Who it's for

Complete beginners, and anyone who uses AI tools daily but has never had the pieces explained. It's also a good refresher before you dive into building with AI, because the same words show up in every tool's documentation and pricing page.

Why we picked it

It's short, calm and comes from IBM's technology team rather than a hype channel. You'll come away able to read an AI product page and understand what it's promising. Pair it with our glossary when a term from the video sticks with you.

Tip: keep a note open while you watch and write one sentence per concept in your own words. If you can explain RAG to a friend afterwards, you've got it.

Tools mentioned

Gemini logoGeminiGeminiFreemium

Gemini is Google's family of multimodal AI chat assistants, accessible through a web interface at gemini.google.com. It lets users converse with an AI model, ask questions, and work with different types of content. The website is built around a chat input area where users type prompts and receive responses. It is aimed at general users, students, professionals, and developers who want an AI assistant integrated with Google's ecosystem. Because the page content retrieved was mostly technical styling code rather than descriptive marketing copy, detailed feature and pricing information could not be fully confirmed from this page alone.