Open Source vs Closed AI Models — Why It's a Big Deal

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-builders-foundations-AI-and-open-vs-closed-models-teen

1. What makes a model 'open source' in the context of artificial intelligence?

   1. The model automatically shares your data with other users
   2. The model is created by volunteers without any funding
   3. The model is available for free but cannot be modified
   4. The model can be downloaded and run on your own computer

2. Which of the following AI models is explicitly described as a closed model in the material?

   1. Llama
   2. DeepSeek
   3. ChatGPT
   4. Mistral

3. What is a key advantage of running an open source AI model on your own server?

   1. You have full control over your data and how the model is used
   2. You can use it without any technical knowledge
   3. You never need to pay for electricity
   4. The model will automatically improve itself

4. Why might someone choose to use a closed AI model instead of an open one?

   1. Closed models are easier to use without technical setup
   2. Closed models always produce more accurate answers
   3. Closed models are always free to use
   4. Closed models allow you to modify the underlying code

5. What are 'model weights' in the context of AI?

   1. The numerical values that determine how an AI model makes decisions
   2. The limits placed on what an AI can say
   3. The copyright laws that protect AI companies
   4. The physical components of a computer that run AI

6. According to the material, what is one reason open source AI models are valuable for learning?

   1. They are created by school teachers
   2. They require no internet connection to function
   3. They show you exactly how the AI works inside
   4. They are always more advanced than closed models

7. What does it mean that closed AI models create 'dependency' on a company?

   1. The company owns the data you input
   2. You become legally obligated to use that AI forever
   3. You rely on the company to keep the model running and decide its future
   4. You must use the company's brand of computer

8. What is a key reason the choice between open and closed AI matters for society?

   1. It controls who has access to the internet
   2. It decides whether AI will exist in the future
   3. It determines which company will win the AI race
   4. It affects who controls the technology and who can build with it

9. If a company suddenly shuts down its closed AI service, what happens to users?

   1. They lose access to that AI tool with no backup option
   2. They can sue the company for damages
   3. They can simply download the code and keep running it
   4. They can force the government to reopen it

10. Based on the material, which type of AI model would be most useful for someone building a custom application?

    1. A closed model that charges per word
    2. Any model that is popular on social media
    3. A closed model accessed through an API
    4. An open model they can modify and integrate

11. Why might a business choose to use a closed AI model for customer service?

    1. Because it requires no technical team to set up and maintain
    2. Because it automatically complies with all laws
    3. Because it is always more accurate than open models
    4. Because it cannot be used for harmful purposes

12. What is a 'black box' in the context of AI?

    1. A security feature that protects AI systems
    2. A type of AI that only works at night
    3. A system where you cannot see or understand how it works
    4. A physical container for AI computers

13. According to the material, what is a reason both open and closed AI matter?

    1. Open shows how it works; closed shows the frontier
    2. They are actually the same thing
    3. They both cost the same amount of money
    4. One is always better than the other

14. What is required to run most open source AI models on your own computer?

    1. A powerful enough computer, especially a good GPU
    2. Nothing — they run on air
    3. A subscription to a cloud service
    4. A government license

15. What is one reason someone might want to study the code of an open source AI model?

    1. To learn how machine learning actually works
    2. To find security vulnerabilities to exploit
    3. To copy it exactly and sell it as their own
    4. To delete the model from the internet