Zero to One: Notes on Startups, or How to Build the Future · chapter 12 · member edition

Man and Machine / Seeing Green

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

Claims checked in this chapter (9)

needs context⚠ re-audit pending
Computers complement humans, not substitute for them. The future is about human-computer collaboration, not replacement.
needs context⚠ re-audit pending
Cleantech companies collectively lost massive amounts of capital (Thiel claims investors poured more than $50 billion into cleantech). Most failed because they couldn't answer Thiel's '7 questions' about engineering, timing, monopoly, people, distribution, durability, and secrets.
contested⚠ re-audit pending
Solyndra exemplifies cleantech failure: received hundreds of millions in government loans, built on flawed technology, and went bankrupt.
needs context⚠ re-audit pending
Palantir exemplifies human-computer collaboration: analysts combined with Palantir's software caught fraud and mapped insurgent networks better than either humans or software alone.
contested⚠ re-audit pending
Globalization is substitution (cheap labor replaces workers) while technology is complementation (computers empower workers). The two forces work differently on employment.
needs context⚠ re-audit pending
Every successful company must answer seven questions: engineering (10x better tech), timing, monopoly, people, distribution, durability, and secrets.
needs context⚠ re-audit pending
Tesla answered all seven questions correctly: breakthrough technology (other automakers licensed Tesla's batteries), small market start ($109K Roadster), direct-to-consumer distribution, engineer-founder (Musk), and cleantech-as-fashion secret.
contested⚠ re-audit pending
Solar panel efficiency improved only ~25-40% (from ~16% to ~22%), nowhere near the 10x improvement Thiel says is needed for breakthrough technology.
contested⚠ re-audit pending
Cleantech founders lacked technical backgrounds -- Thiel characterizes them as 'men in suits' rather than genuine technologists who 'wear T-shirts and jeans.'