Quiz — Chapter 21: The Road Ahead

Multiple Choice

1. Multimodal AI refers to:

a) AI that can only process text b) AI systems that can process and generate across multiple types of data (text, image, audio, video) c) AI that runs on multiple computers simultaneously d) AI that is available in multiple languages

2. AI agents differ from current prompt-response AI systems primarily because they:

a) Use larger datasets b) Can plan and execute multi-step tasks with greater autonomy c) Are always connected to the internet d) Are only available to enterprise customers

3. Scenario planning is useful for thinking about AI's future because:

a) It guarantees accurate predictions b) It identifies a single most-likely outcome c) It explores multiple plausible futures, helping us prepare for different possibilities d) It eliminates uncertainty about technological change

4. The concept of "durable frameworks" in this chapter refers to:

a) Hardware that lasts longer than software b) AI systems designed to operate for decades c) Analytical tools and questions that remain useful regardless of how specific technologies evolve d) Legal frameworks that cannot be amended

5. Technological determinism is the belief that:

a) Technology should be used to determine government policy b) The development and impact of technology is inevitable and cannot be shaped by human choices c) Only technologists should make decisions about technology d) Technology always produces positive outcomes

6. The chapter argues that the most important outcome of AI literacy is:

a) The ability to code AI systems b) The ability to predict exactly which AI technologies will succeed c) The ability to ask the right questions about any AI system, even those not yet invented d) The ability to avoid using AI entirely

7. Which of the following is described as a "wild card" that could change AI's trajectory?

a) Continued improvement of existing transformer models b) A new AI architecture that fundamentally replaces transformers c) Gradual adoption of AI in healthcare d) Incremental improvements in RLHF techniques

8. In the "muddle through" scenario, AI development is characterized by:

a) Catastrophic failure that turns public opinion against AI b) Rapid progress that solves all major societal problems c) Continued advancement with benefits and harms distributed unevenly d) Complete stagnation of AI research

9. The chapter's central argument about the citizen's role in AI's future is that:

a) Citizens should leave AI governance to technical experts b) Citizens cannot meaningfully influence AI's development c) The future of AI is shaped by human choices, and informed citizen participation makes those choices better d) Citizens should refuse to use AI systems

10. The FACTS Framework, threshold concepts, and recurring themes are collectively described as:

a) Outdated tools that need to be replaced with newer frameworks b) Technical skills that require coding ability to apply c) Components of a personal AI literacy toolkit designed to remain useful as technology evolves d) Marketing tools used by AI companies

True or False

11. The chapter predicts that one specific scenario for AI's future is most likely. True / False

12. AI agents raise new accountability challenges because they take autonomous actions, not just generate responses. True / False

13. The chapter argues that AI literacy is a one-time achievement that, once acquired, never needs updating. True / False

14. The "augmentation era" scenario assumes that governance effectively catches up to technology. True / False

15. According to the chapter, AI expertise and AI literacy are the same thing. True / False

Short Answer

16. Explain why the chapter compares AI's current trajectory to the trajectory of the internet. What parallel does it draw, and what lesson does it suggest?

17. The chapter argues that "the less visible AI becomes, the more important AI literacy becomes." Explain this claim in 2–3 sentences, giving one concrete example.

18. Choose one of the four anchor examples (ContentGuard, MedAssist AI, Priya's Semester, CityScope Predict) and describe how your understanding of it has evolved from Chapter 1 to Chapter 21. What do you see now that you did not see at the beginning?

Scenario-Based

19. You are a newly elected city council member. A company approaches your city with an AI-powered system that promises to reduce emergency response times by 40%. The system uses historical data to predict where emergencies will occur and pre-positions ambulances accordingly. Using the AI literacy toolkit developed in this book, identify five specific questions you would ask before voting on whether to adopt this system. For each question, identify which chapter or framework it draws on. (Answer in 200–250 words.)

20. It is 2035. AI has continued to advance significantly. You are explaining to a younger person why they should develop AI literacy. Drawing on this chapter and the full course, write a 150-word argument for why AI literacy will still matter in a world where AI is even more capable and pervasive than it is today.

Answer Key

  1. b
  2. b
  3. c
  4. c
  5. b
  6. c
  7. b
  8. c
  9. c
  10. c
  11. False — The chapter deliberately presents three scenarios without claiming one is most likely, because the purpose is to prepare for multiple possibilities.
  12. True
  13. False — The chapter emphasizes that AI literacy is an ongoing practice, not a one-time achievement, because the technology continues to evolve.
  14. True
  15. False — AI literacy is the ability to critically evaluate and engage with AI systems; AI expertise requires years of specialized technical study. The chapter argues that literacy, not expertise, is what most people need.
  16. Sample: The chapter draws a parallel between the internet's evolution and AI's trajectory, noting that the internet's early promise of democratic participation gave way to surveillance capitalism, misinformation, and privacy erosion — outcomes shaped by specific policy choices. The lesson is that AI's future is similarly malleable, and the choices we make now will determine whether AI produces equitable benefits or concentrated harms.
  17. Sample: As AI becomes embedded invisibly in tools people already use — email, search engines, medical records — users interact with AI without realizing it. This makes it harder to evaluate AI's influence on decisions affecting them. For example, if an AI system silently prioritizes some job applicants over others within a standard recruiting platform, applicants who are unaware of the AI cannot question its decisions.
  18. Open-ended; strong answers will trace a specific evolution from surface understanding to multi-layered analysis.
  19. Open-ended; strong answers will identify questions drawing on FACTS Framework (Ch.1), data/bias (Ch.4, 9), accountability (Ch.13, 17), and safety/alignment (Ch.20).
  20. Open-ended; strong answers will argue that more capable AI makes literacy more important, not less, because the stakes of uninformed engagement rise with AI's power.