Chapter 6 Further Reading: Who Builds These Systems?

These recommendations are organized by type of source and reading difficulty. All works cited were published or accessible as of 2025.

Primary Accounts and Journalism

1. The Wall Street Journal, "The Facebook Files" series (September–October 2021) The original reporting that broke the Facebook Papers story, based on documents provided by Frances Haugen. The Journal's six-part series published findings on Instagram and teen mental health, the vaccine misinformation amplification problem, and internal knowledge of the recommendation rabbit hole effect. Available in full on the Journal's website. Essential reading for anyone studying the gap between platform public commitments and internal knowledge. Note that the Journal's reporting was later supplemented by a broader consortium of outlets publishing simultaneously with Haugen's congressional testimony. Start here, then cross-reference with The Guardian and NPR's coverage for the full breadth of the Papers' findings.

2. Antonio García Martínez, Chaos Monkeys: Obscene Fortune and Random Failure in Silicon Valley (Harper, 2016) A memoir by a former Facebook product manager who worked on the company's advertising targeting systems. Martínez writes in a voice that is self-consciously provocative, and his account should be read with an awareness of his rhetorical choices. But the inside view of Facebook's product culture — how decisions were made, what was measured, what mattered and what didn't — is detailed and illuminating. The advertising system chapters are particularly relevant to understanding how the business model shapes everything else. The book was controversial on publication and remains so; reading it alongside critiques of García Martínez's persona will give you a fuller picture.

3. Sheera Frenkel and Cecilia Kang, An Ugly Truth: Inside Facebook's Battle for Domination (Harper, 2021) A deeply reported account by two New York Times technology journalists of Facebook's evolution from startup to global platform, with a particular focus on the period from 2016 to 2021 — the Cambridge Analytica scandal, the 2016 election misinformation controversy, the internal debates about content moderation, and the run-up to Haugen's disclosure. Based on interviews with hundreds of current and former employees. Provides essential context for understanding the organizational culture and leadership dynamics that shaped the decisions documented in the Facebook Papers. One of the best single-volume accounts of platform company culture available.

4. Alex Kantrowitz, "I Have Blood on My Hands: A Whistleblower Says Facebook Ignored Global Political Manipulation" (BuzzFeed News, September 2020) The original publication of Sophie Zhang's internal memo. Available free online. A direct primary source — Zhang's own words about what she found, what she reported, what happened, and how she evaluated her own complicity. Shorter and more accessible than book-length accounts. Read this alongside later reporting on what happened after the memo's publication to trace whether and how it changed anything at Facebook.

Academic and Research Sources

5. Safiya Umoja Noble, Algorithms of Oppression: How Search Engines Reinforce Racism (NYU Press, 2018) A rigorous examination of how Google's search algorithm design reflected and reinforced racial biases. Noble's analysis of specific search results and their implications for how different populations are represented and characterized is detailed and evidence-based. The book is essential for understanding how demographic homogeneity in engineering teams produces design outcomes that systematically disadvantage underrepresented groups — not through intentional discrimination but through the reproduction of existing cultural assumptions in algorithmic systems. Accessible to non-specialists; well-cited for further investigation.

6. Emily Bender, Timnit Gebru, Angelina McMillan-Major, and Shmargaret Shmitchell, "On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?" (FAccT 2021) The paper at the center of Timnit Gebru's firing. Available free online through ACM Digital Library and on author websites. Reading the paper itself — rather than relying on descriptions of it — is instructive on multiple levels: the substance of its arguments (which are technically sound and have aged well), its tone (rigorous but accessible), and the specific features of the argument that challenged Google's commercial interests. After reading the paper, compare it with Google's public responses and consider what a substantive technical rebuttal would need to address.

7. Albert Bandura, "Moral Disengagement in the Perpetration of Inhumanities" (Personality and Social Psychology Review, 1999) The foundational academic source for the moral disengagement framework used throughout this chapter. Available through academic databases. The article is dense but carefully written, and the specific mechanisms Bandura identifies — moral justification, euphemistic labeling, advantageous comparison, displacement of responsibility, diffusion of responsibility, dehumanization — are precisely applicable to platform company culture. Reading the original source alongside the chapter's application of it will deepen your understanding of both.

8. Batya Friedman, Peter H. Kahn, Jr., and Alan Borning, "Value Sensitive Design and Information Systems" (in The Handbook of Information and Computer Ethics, 2008) The foundational text on value-sensitive design, a methodology for incorporating human values into the engineering design process. Available through academic databases. The methodology offers a formal vocabulary for engineers who want to raise ethical concerns in technical settings — a vocabulary that can be heard by colleagues trained in quantitative reasoning. Understanding the VST approach is useful for thinking about what a wellbeing review process, like the one Aisha Johnson is drafting in the Velocity Media narrative, might actually look like in practice.

Investigative Journalism and Long-form Reporting

9. The Markup, ongoing platform accountability coverage (themarkup.org) The Markup is a nonprofit investigative newsroom focused on how technology affects society, with particular attention to algorithmic systems. Their "Citizen Browser" project tracked Facebook's news feed content in real time, providing quantitative evidence about what the algorithm was actually amplifying. Their "Google the Giant" series documented Google's self-preferencing in search results. Rigorous, technically sophisticated, and freely accessible. Subscribing to their newsletter is an efficient way to stay current on platform accountability journalism.

10. Julia Angwin, Jeff Larson, Surya Mattu, and Lauren Kirchner, "Machine Bias" (ProPublica, May 2016) A landmark investigation into COMPAS, the algorithmic risk assessment tool used in criminal sentencing, which ProPublica found was racially biased in specific, measurable ways. Available free on ProPublica's website. While COMPAS is a criminal justice system rather than a social media platform, the investigation established a model for evidence-based reporting on algorithmic harm that influenced subsequent coverage of platform technology. The methodology — obtaining algorithmic outputs for a representative sample and analyzing them for differential impact — is now a standard tool in tech accountability journalism.

11. Kashmir Hill, New York Times technology reporting (2019–present) Kashmir Hill has produced some of the most methodologically rigorous journalism about surveillance, facial recognition, and platform data practices. Her reporting on Clearview AI (a facial recognition company that scraped billions of images from social media) and on what happens to people who are misidentified by facial recognition algorithms is directly relevant to themes in this chapter. Her work demonstrates what investigative journalism into platform harms looks like when conducted with technical depth and careful documentation. Searchable in the Times archives.

Broader Context

12. Frank Pasquale, The New Laws of Robotics: Defending Human Expertise in the Age of AI (Harvard University Press, 2020) Pasquale is a law professor who has written extensively about algorithmic accountability, and this book addresses the question of what legal and institutional frameworks are needed to govern AI systems effectively. Particularly relevant to this chapter's discussion of whether self-regulation can work: Pasquale's analysis of the structural conflict of interest in corporate AI ethics work provides a rigorous legal and institutional account of the problem the Gebru case illustrates. More demanding than the journalism recommendations but important for understanding the regulatory dimension.

13. Philip Zimbardo, The Lucifer Effect: Understanding How Good People Turn Evil (Random House, 2007) Zimbardo's book-length examination of how ordinary people come to participate in harmful systems, developed from his work on the Stanford Prison Experiment and his subsequent analysis of Abu Ghraib. The book provides the psychological and institutional context for understanding moral disengagement in large organizations. Chapter 12 ("Investigating Social Dynamics: Power, Conformity, and Obedience") and Chapter 15 ("Putting the System on Trial") are most directly relevant to the platform culture analysis in this chapter. Read critically — Zimbardo's conclusions have been contested on methodological grounds — but the framework remains influential.

14. Ben Green, "The Impossibility of Fairness: A Formalization of Tradeoffs Between Sensitivity, Specificity, and Predictive Value in Algorithmic Decision-Making" (Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency) A more technical academic paper demonstrating that multiple common definitions of algorithmic fairness are mathematically incompatible with each other — you cannot simultaneously satisfy all of them — in any system where base rates differ between groups. Relevant to this chapter because it illustrates the kind of ethics research that can be done rigorously within technical constraints: it does not require access to proprietary data, it does not challenge commercial interests directly, and it produces findings that are precisely applicable to system design. Understanding what this kind of research can and cannot do helps clarify the limits and possibilities of academic ethics work on platform systems. Available free through the ACM Digital Library.

A Note on Finding Current Coverage

Platform accountability journalism moves quickly. By the time you read this, significant new reporting, regulatory actions, and whistleblower disclosures may have occurred that are directly relevant to the themes of this chapter. The following sources reliably cover these developments:

  • The Verge for platform company culture and internal reporting
  • Wired for technical depth on AI and platform systems
  • The Guardian's Technology section for international coverage of platform regulatory developments
  • MIT Technology Review for research-adjacent coverage of AI ethics and safety questions
  • Rest of World for coverage of how platform decisions affect communities in the Global South — particularly relevant given the Sophie Zhang case's documentation of Facebook's underinvestment in non-Western markets