Get Certified for Free Browse courses

Using AI · Level 1 of 5 · Intro

What AI assistants are

What a chat assistant does, what it cannot do, and why it sometimes makes things up.

Study notes

  1. AI chat assistants are built on large language models (LLMs). An LLM is trained on large amounts of text and generates a reply by predicting likely next words.
  2. Because the model predicts plausible text, it can state wrong facts confidently. These errors are often called hallucinations.
  3. Many assistants have a training cutoff date. Unless the tool can search the web, it may not know recent events.
  4. Assistants are good at drafting, summarizing, rewording, brainstorming, and explaining. They are weaker at exact facts, math without a calculator tool, and anything they cannot see.
  5. You stay responsible for what you send, publish, or decide based on AI output.

Quiz: 6 questions

Choose one answer for each question, then check your answers. Score 5 of 6 to pass. Then enter your name and email to download a PDF certificate and have a copy emailed to you. Answers are graded in your browser.

1. What is a large language model mainly trained to do?

2. An AI assistant gives you a confident answer with a made-up statistic. This is usually called:

3. Why might an assistant not know about last week's news?

4. Which task is an AI assistant usually best at?

5. Who is responsible for an email you send that an AI drafted?

6. Which statement is true?

About this certificate: Certificates are issued by Get Certified for Free for passing a free online knowledge quiz. They are not accredited, licensed, government, academic, or professional credentials and do not confer a license. Read the full terms.