CBR: The AI Con. Chapter 1: The Hype of Thinking Machines

Emily M. Bender and Alex Hanna. The AI Con. How To Fight Big Tech’s Hype and Create the Future We Want. Harper Books. 2025.

Chapter Reviewed: Introduction: The Hype of Thinking Machines.

Review by: Alp Eren Pirli, PhD Candidate in English, Indiana University Bloomington

The title of Emily M. Bender and Alex Hanna’s 2025 book, The AI Con: How to Fight Big Tech’s Hype and Create the Future We Want, positions their argument as uncompromisingly polemic. A polemic that cannot define what it opposes would fall flat on its face; hence, the primary objective of The AI Con’s first chapter, “An Introduction to AI Hype,” is to elucidate what “AI,” and the hype around it, exactly are.

Bender and Hanna define hype as “the aggrandizement of some person, artifact, technology, or technique that you, the consumer, absolutely need to buy or invest in as early as possible, lest you miss out on entertainment or pleasure, monetary reward, return on investment, or market share.” AI marketers—“‘AI’ is a marketing term,” Bender and Hanna assert—weaponize hype by framing the discourse on AI around scary-sounding, sci-fi-esque language. It is difficult to provide a counterargument to this claim; anyone who spends their time online is bound to see some interesting characters swooning about artificial general intelligence (AGI)—a sort of übermensch ex machina whose cognitive capabilities surpass the human mind.

The hype around AGI is discussed in the next chapter; in chapter 1, Bender and Hanna’s attention is on the “mathy-looking” metrics of p(doom) and p(hope)—the focal point of the December 6, 2023 Artificial Intelligence Insight forum organized by the then-Senate Majority Leader Chuck Schumer. “Mathy-looking,” but not quite mathy: though these terms that look like probability metrics sound important—AI could either spell our ruin, or bring about a new stage in human evolution!—their “estimates are based in fantasy rather than data or empirical fact.” All the while, “techno-enabled doom” is nothing if not real for marginalized individuals, examples of which range from Black people being arrested due to errors made by facial recognition models, to deepfake porn apps that high school students use to sexually harass their bullying victims. Although software marketed as “AI” underlie both of these instances (and more), “harms befalling real people” are not the concern of p(doom). Rather, the rhetoric of p(doom) serves to distract us from the actual danger brought about by AI.

p(doom), as well as its positive counterpart projecting “a fully automated paradise” in the future, is all part of the AI con. One instills in the public the fear of total destruction brought about by an unchecked technology, the other, the fear of being left in the dust of those who adopted a groundbreaking invention. Both are equally problematic, in that both work to sell  “a bill of goods … to line someone’s pockets.” This rhetoric around AI is “all bullshit” claim Bender and Hanna, proposing to dismantle it “piece by piece.” It is convincing that they will do so; they state their credibility clearly: “The two of us have spent the past few years attacking AI hype and puzzling around what’s behind it. Emily is a linguist who’s gained recognition for raising ethical issues in the development of language technology. Alex is a sociologist who formerly worked on Google’s Ethical AI team and thinks a lot about how technology and society interact.” It makes sense for the expertise of both authors to be established simultaneously; doing so signals an interdisciplinary approach that tackles both the technical and the societal concerns surrounding AI.

Despite this, some readers may find The AI Con’s tone too dismissive and snarky, a charge which Bender and Hanna admit to. Even so, it must be noted that this tone is accompanied by one that is sensitive and sincere, exemplified in this chapter by the accentuation of the actual AI doom scenarios faced by disempowered individuals. It is the intertwining of these two tonal registers that make The AI Con an exceptionally strong polemic.

This polemic does not have the sole purpose of railing against the rhetoric surrounding AI. The snapshot of the six chapters that follow, which concludes this chapter, ends with the “hope” of showing the reader “how to resist the urge to be impressed, how to spot AI hype in the wild, and how to take back some ownership in our technological future.” What is displayed in this introductory chapter promises that this hope will be realized.