Digital Mind 🧠

🪵 The Threshold

Emily M. Bender, Alex Hanna — The AI Con- How to Fight Big Tech's Hype and Create the Future We Want

5 min read

Emily M. Bender, Alex Hanna — The AI Con- How to Fight Big Tech’s Hype and Create the Future We Want

Yield

22 of 1242 paragraphs kept (1.8%). Target for mining: 30%–50% — ⚠️ BELOW target.

Pre-Assessment

Emily M. Bender is a prominent linguistics professor at the University of Washington known for her critical work on Large Language Models (co-author of the seminal ‘Stochastic Parrots’ paper). Alex Hanna is a sociologist and Director of Research at the Distributed AI Research Institute (DAIR), specializing in the political economy of AI. Together, their expertise is exceptionally current and grounded in critical AI studies and data justice. Given the book’s title and authors, the yield will likely be a dense mix of critical frameworks (concept mining) and concrete mechanisms of how hype is manufactured and labor is exploited (fact mining), directly aligning with the user’s focus on the ‘AI con’.

Reading Goal

Prioritizing: Causal Relationships (To trace how AI causes harm and creates false expectations.), Contrarian or Surprising Claims (To uncover hidden deceptions or surprising manipulation tactics.), Factual Claims (To examine evidence of AI deception and data extraction.), Connection Points (To link AI business models to con artist behavior.), Definitional Content (To understand how AI systems are defined versus how they operate.) De-prioritizing: Craft and Method (Less relevant to the persuasive/deceptive tactics of AI.), Inspiration and Speculation (Focus is on current deceptive practices rather than speculation.)

Thesis

Documenting hype contours helps inhibit future tech bubbles. · ch. 1

Argument

  1. Hype can mislead public perception of AI technology. — ch. 1
  2. AI hype continues to escalate beyond expectations. — ch. 1
  3. AI marketing manipulates perceptions to generate profits, ignoring real harms. — ch. 2
  4. Language models lack subjectivity and do not understand meaning as humans do. — ch. 3
  5. The general intelligence project has racist roots affecting societal inequalities. — ch. 3
  6. Goldman Sachs claims AI could replace 300 million jobs worldwide. — ch. 4
  7. Researchers relied on GPT-4 about labor market impacts, compromising their findings. — ch. 4
  8. AI reduces job quality by devaluing human labor. — ch. 4
  9. Replacing skilled labor with AI can result in harm, not benefit. — ch. 4
  10. AI tools require hidden labor for data management and content screening. — ch. 4
  11. AI solutions will not improve access to healthcare as claimed. — ch. 5
  12. AI lacks the ability to address deep-rooted social issues effectively. — ch. 5
  13. The absence of citational practice in AI undermines recognition of original creators’ contributions. — ch. 6
  14. AI does not demonstrate creativity; it merely appropriates human contributions. — ch. 6
  15. AI systems are using artists’ work without their consent. — ch. 6
  16. Many AI Doomers profit from the dangers they exaggerate. — ch. 7
  17. AI development is not an inevitable process dictated by technology or necessity. — ch. 7
  18. Current ideologies in AI reflect historical eugenics concepts. — ch. 7
  19. Access systems interfere with users’ information navigational skills. — ch. 8
  20. Emerging technologies can obscure important social responsibilities. — ch. 8
  21. Criminality cannot be determined by facial recognition alone. — ch. 8

Annotation Roundup

Question

✉️