Yudkowsky and Soares frequently employ analogies to explain complex concepts, comparing AI alignment to challenges from nuclear engineering (Enrico Fermi), computer security, and even pre-scientific alchemy. Select two key analogies from the book and analyze how they function to make abstract dangers tangible and persuasive for a non-expert audience.

The authors present real-world figures like Sam Altman, Geoffrey Hinton, and Yann LeCun as key actors in the AI landscape. Analyze how the book uses public figures to dramatize competing views about AI development. What does this approach add to the argument, and what risks does it create for the book’s treatment of institutional responsibility?

Explore the authors’ use of the “grown, not crafted” paradigm as a central metaphor. How does this concept shape their arguments about AI development, engineering knowledge, and technological control? What assumptions about safety and reliability does this metaphor invite readers to question?

Part 2, the story of Sable, functions as a work of speculative fiction embedded within a nonfiction argument. Analyze this section through the conventions of the techno-thriller genre. How do the authors use elements of pacing, suspense, and a non-human protagonist to translate the book’s theoretical warnings into an immediate and visceral threat?

How does the historical parable of Thomas Midgley Jr. and leaded gasoline shape the book’s discussion of technological optimism? What parallels and limits do the comparison establish between past technological disasters and contemporary AI development?

Examine the distinction Yudkowsky and Soares draw between “prediction” and “steering” as the two components of intelligence. How does this framework affect the book’s treatment of AI goals, agency, and risk? What questions does it raise about the relationship between intelligence and human values?

The book uses the history of the Cold War and nuclear risk as both a parallel and a point of contrast. Analyze how the authors invoke figures like Vasily Arkhipov and events like the Cuban Missile Crisis to discuss international cooperation, existential risk, and strategic restraint. Where do the comparisons between nuclear weapons and ASI seem most persuasive, and where do they become more difficult to sustain?

How does the proposal for a globally enforced military halt to AI development shape the reader’s understanding of the problem presented throughout the book? What assumptions about regulation, technological progress, and international cooperation does this proposal depend on?