AI model families: multimodal AI (text + image + audio)

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-builders-model-families-AI-and-multimodal-teen

1. What makes a model 'multimodal'?

   1. It learns faster than other AI models
   2. It can handle and combine information from different formats like text, images, and sound
   3. It can only process text written by humans
   4. It works without internet connections

2. A student takes a photo of a math problem on a whiteboard and asks an AI to solve it. This is an example of:

   1. Multimodal AI using vision capabilities
   2. A video call with AI
   3. A voice interaction
   4. Text-only AI processing

3. Why might showing an AI a picture of a confusing sign be better than typing out a description?

   1. Pictures load faster than text
   2. The AI might misunderstand a description but can read the sign directly
   3. Text descriptions cost more money
   4. AI cannot read text at all

4. Which of these is NOT mentioned as a use case for multimodal AI in the lesson?

   1. Describing a meme to someone who cannot see
   2. Writing code from a sketch
   3. Debugging a website by showing a screenshot
   4. Sending a voice note instead of typing

5. What does it mean when the lesson says AI can 'see' images?

   1. The AI has eyes like a human
   2. The AI uses computer vision to analyze pictures
   3. The AI stores every image it sees forever
   4. The AI turns images into text files

6. A user sends a voice note to an AI assistant instead of typing. What capability is being used?

   1. Image generation
   2. Text-to-speech
   3. Video processing
   4. Audio input or speech recognition

7. The lesson says: 'If you can show it, you don't have to describe it.' What is the main advantage of this approach?

   1. Descriptions are always wrong
   2. It saves the AI memory
   3. It makes the AI work faster
   4. Showing avoids losing information in translation—you get exactly what's in the image

8. Which of these inputs would a multimodal AI definitely be able to process?

   1. A screenshot of an error message and a question about it
   2. Only typed text
   3. Only photos without any accompanying text
   4. Only voice recordings

9. Your grandma cannot see a funny meme online. How could multimodal AI help?

   1. By describing what the meme shows using text or speech
   2. By sending the meme to her phone
   3. By making the meme larger
   4. By fixing her computer screen

10. What would happen if you tried to show a photo of a recipe to a text-only AI?

    1. It could not 'see' the image and would ask you to type the recipe
    2. It would work exactly the same as a multimodal AI
    3. It would automatically convert the image to text
    4. It would refuse to respond

11. The lesson mentions GPT-5, Claude, and Gemini as examples of multimodal models. What do these represent?

    1. Different programming languages
    2. Different companies that make AI assistants
    3. Different types of computer hardware
    4. Different file formats

12. You see a chart with confusing data on it. What could you do with a multimodal AI that you couldn't do with a text-only AI?

    1. Ask it to make the chart bigger
    2. Print the chart
    3. Delete the chart
    4. Show it the chart and ask what the data means

13. Why is it useful that multimodal AI can 'hear' audio?

    1. It can record your voice
    2. It can play music for you
    3. It can sing songs
    4. It can understand spoken questions without you typing

14. A developer shows a multimodal AI a screenshot of a broken website button. What is the likely goal?

    1. To save the screenshot
    2. To have AI help find and fix the problem
    3. To test if the AI can see colors
    4. To have AI describe the screenshot to a blind person

15. What is computer vision?

    1. A type of video game
    2. The ability of AI to analyze and understand images
    3. A camera that computers wear
    4. A way to draw pictures with computers