Lesson 1 of 6

What people mean when they say "AI"

"AI" is a broad label. Here's what it usually means today, and the three kinds you'll actually meet in a business.

3 min readBy Michael Alos

"AI" gets used for everything from a spam filter to a robot in a science fiction film. That makes it hard to know what anyone actually means. Let's sort it out.

Artificial intelligence is a broad name for computer systems that do things which normally need human judgment: recognizing a face, spotting fraud, understanding a question, writing a paragraph.

Three layers, from broad to specific

  1. AI is the umbrella term for all of it.
  2. Machine learning is the most common way AI gets built today. Instead of a programmer writing every rule, the system learns patterns from a large number of examples. Show it thousands of spam emails and thousands of normal ones, and it learns to tell them apart.
  3. Generative AI is machine learning that creates something new: text, images, audio or video. ChatGPT, Claude, Gemini and Copilot are all generative AI.

When people talk about AI in a business conversation today, they almost always mean generative AI. That's what this course focuses on.

You've been using AI for years

AI isn't new. You've relied on it for a long time without the label:

  • Your email's spam filter
  • The fraud alert from your bank when a strange charge appears
  • Traffic predictions in your maps app
  • Product recommendations when you shop online
  • Autocomplete when you type a text message

Those systems sort and predict. What changed recently is that AI got good at writing and conversation. That's what made it useful for everyday business work like emails, quotes, summaries and customer questions.

The three kinds you'll meet in a business

  1. AI chat assistants. You type a request, and it answers or drafts something. ChatGPT, Claude, Gemini and Copilot are the best known. This is where most people start.
  2. AI inside software you already use. Your email, office suite, CRM, accounting software and website builder are adding AI features: drafting replies, summarizing meetings, suggesting next steps.
  3. AI connected into your workflows. AI becomes one step in a process. A new inquiry arrives, AI drafts a reply, and a person approves it before it's sent.

Most people start with the first kind. Over time, small businesses usually get the most value from the second and third, because the AI is working inside the jobs that already happen every day.

What AI isn't

AI isn't a person, and it doesn't understand your business the way you do. It only knows what it was trained on and what you tell it. It can also be wrong while sounding completely sure of itself. Lesson 4 covers that in detail.

A useful way to think about today's AI: a fast, well-read assistant who has never worked at your business, and who sometimes makes things up with total confidence.

Try this (5 minutes)

Open any AI chat assistant. The free version is fine.

  1. Describe your business in two sentences.
  2. Then ask: "Explain what my business does to a new employee in three sentences."
  3. Read the answer. What did it get right? What did it assume or invent?

That gap between what you told it and what it filled in is the most important thing to notice about AI.

Common mistakes

  • Treating every product labeled "AI" as the same thing. A spam filter and a writing assistant are very different tools.
  • Assuming an AI assistant knows facts about your business that you haven't given it.
  • Waiting for AI to "settle down" before trying it. The basics are already stable enough to be useful.

Check yourself

What's the difference between machine learning and generative AI?

Machine learning learns patterns from examples so it can sort or predict. Generative AI uses learned patterns to create something new, such as text or images.

What are the three kinds of AI you're likely to meet in a business?

Chat assistants, AI features inside the software you already use, and AI connected into your workflows.

Does an AI assistant know how your business works?

No. It only knows what it was trained on and what you tell it or give it access to.

Where could AI give your business the most back?

Answer thirteen questions and get your biggest opportunities ranked, with a rough figure for each. Free, about four minutes, no email required.

Take the AI Opportunity Check →