Further Reading — Chapter 1

What Is Artificial Intelligence? (Separating Fact from Fiction)

Tier 1: Verified Sources (peer-reviewed or established institutional publications)

1. Nils J. Nilsson, The Quest for Artificial Intelligence: A History of Ideas and Achievements (Cambridge University Press, 2010). A comprehensive history of AI from its origins through the early 2010s, written by one of the field's pioneers. Nilsson traces the intellectual lineage of AI concepts from logic and philosophy through the Dartmouth workshop and beyond. Excellent for understanding why AI is a spectrum of techniques, not a single invention. Chapter 1 of this textbook draws on the historical framing that Nilsson develops across several hundred pages.

2. Meredith Broussard, Artificial Unintelligence: How Computers Misunderstand the World (MIT Press, 2018). Broussard, a data journalist and computer scientist, argues that "technochauvinism" — the belief that technology is always the solution — leads us to overestimate what AI can do and underestimate the importance of human judgment. Particularly relevant to this chapter's discussion of the gap between AI capability and AI understanding. Accessible and well-sourced.

3. Ziad Obermeyer, Brian Powers, Christine Vogeli, and Sendhil Mullainathan, "Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations," Science 366, no. 6464 (2019): 447–453. The landmark study that found a widely used healthcare algorithm systematically underestimated the health needs of Black patients by using healthcare spending as a proxy for health need. Essential reading for understanding how AI systems can produce biased outcomes even without explicit discriminatory intent. Referenced in Case Study 1.

4. Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach, 4th edition (Pearson, 2020). The standard academic textbook in the field. While it is a technical text, the first three chapters — covering what AI is, the history of the field, and the philosophy of AI — are accessible to motivated non-technical readers. Useful as a reference for precise definitions of key terms used throughout this textbook.

5. Kate Crawford, Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence (Yale University Press, 2021). Crawford examines AI not as a purely technical phenomenon but as a system embedded in networks of power, labor, and natural resource extraction. Her framework for analyzing "who benefits and who is harmed" closely parallels the Consequences question in our FACTS Framework. A challenging but rewarding read that reveals dimensions of AI most introductory treatments overlook.

Tier 2: Attributed Sources (reputable journalism, institutional reports, expert commentary)

6. Julia Angwin, Jeff Larson, Surya Mattu, and Lauren Kirchner, "Machine Bias," ProPublica (May 23, 2016). The investigative report that brought algorithmic bias into mainstream public discourse. The team analyzed COMPAS, a recidivism-prediction tool used in the U.S. criminal justice system, and found that it was significantly more likely to falsely flag Black defendants as high-risk than white defendants. Directly relevant to the CityScope Predict anchor example. Available free online.

7. AI Now Institute, Annual Reports (2017–present). New York University. The AI Now Institute publishes annual reports surveying the social implications of AI across sectors including healthcare, criminal justice, education, and employment. These reports are written for a policy-literate audience and provide up-to-date overviews of AI governance issues. Available at ainowinstitute.org.

8. Arvind Narayanan, "How to Recognize AI Snake Oil" (presentation and working paper, Princeton University, 2019; expanded into AI Snake Oil with Sayash Kapoor, Princeton University Press, 2024). Narayanan, a computer scientist at Princeton, provides a practical framework for distinguishing legitimate AI capabilities from overhyped or fraudulent claims. The original presentation is freely available online and is an excellent companion to section 1.6 of this chapter. The expanded 2024 book with Kapoor deepens the analysis substantially.

9. McKinney et al., "International Evaluation of an AI System for Breast Cancer Screening," Nature 577 (2020): 89–94. The study referenced in Case Study 2 that demonstrated AI matching or exceeding radiologist performance in breast cancer detection from mammograms. Worth reading alongside the case study's discussion of what "superhuman performance" actually means in context and what the headline coverage omitted.

10. Timnit Gebru, Jamie Morgenstern, Briana Vecchione, Jennifer Wortman Vaughan, Hanna Wallach, Hal Daume III, and Kate Crawford, "Datasheets for Datasets," Communications of the ACM 64, no. 12 (2021): 86–92. Proposes a standardized documentation framework for machine learning datasets — essentially a "nutrition label" for training data. Directly relevant to the Training question in the FACTS Framework. If you are interested in how to make AI systems more transparent, start here.

How to Use This List

• Start with Broussard (#2) if you want an engaging, non-technical entry point that challenges assumptions about AI.
• Start with the ProPublica piece (#6) if you prefer investigative journalism and want to see AI bias in a concrete, real-world case.
• Start with Narayanan (#8) if you want practical tools for evaluating AI claims — it pairs naturally with the FACTS Framework.
• Save Russell and Norvig (#4) for later in the course if you develop an interest in the technical foundations.