So what
Chapters 12-13 contain the book’s most time-sensitive claims, and several have been challenged by events since 2014. The human-computer complementarity thesis was reasonable when written: Makela & Stephany (2024) found complementarity effects 1.7x larger than substitution across 12 million job vacancies. But the St. Louis Fed (2025) documents a ~20% decline in entry-level software and customer service employment since ChatGPT’s release. Anthropic’s Economic Index shows 6-16% employment drops for workers aged 22-25 in AI-exposed roles. AI complements senior workers but increasingly substitutes for juniors – Thiel’s binary framing was always too simple. The cleantech critique was factually grounded (MIT confirms $25B+ in VC losses, Solyndra’s $535M default is verified by DOE Inspector General) but used to dismiss an entire sector that subsequently thrived: solar costs dropped 75%, renewables overtook coal globally, and clean energy investment reached $2.2 trillion in 2025. The DOE loan program that funded Solyndra also funded Tesla ($465M, repaid 9 years early) and generated a net profit for taxpayers – making Thiel’s cherry-picking of Solyndra particularly egregious. The seven questions framework is useful for startup evaluation but unvalidated as a scientific framework. Tesla’s point-by-point answers all check out, but retrospective pattern-matching to proven successes is easy.
Verdict
MixedBag