AI Book Review: How Progress Ends: Technology, Innovation, and the Fate of Nations by Carl Benedikt Frey
- Johan Steyn

- 17 minutes ago
- 5 min read
A sweeping history of innovation that asks whether the age of rapid progress is ending – and what that means for AI, growth and global power.

Audio summary: https://youtu.be/1iJ9b2Ob9GI
Follow me on LinkedIn: https://www.linkedin.com/in/johanosteyn/
I read a great many books on technology, economics and artificial intelligence, partly because they inform my work, but also because I am trying to understand the world our children will grow up in. Much of the current AI conversation assumes that progress is inevitable: models will keep improving, productivity will surge, and growth will simply follow. Carl Benedikt Frey’s How Progress Ends: Technology, Innovation, and the Fate of Nations is a deliberate challenge to that story. He takes us on a thousand-year journey through technological change to ask a difficult question: what if progress stalls – or even goes into reverse?
CONTEXT AND BACKGROUND
Frey is well known for his earlier book The Technology Trap and for his work on automation and the future of work at Oxford. In How Progress Ends, published by Princeton University Press in 2025, he widens the lens dramatically. He argues that economic and technological progress have been the exception, not the rule, over most of human history. Periods of rapid innovation – from medieval China to Britain’s Industrial Revolution and America’s rise – were fragile and contingent, and many ended in stagnation when institutions failed to adapt.
The book weaves together history, economics and political economy to explore why some societies manage to harness new technologies while others choke them off. Frey is particularly interested in the balance between decentralised experimentation and centralised control. Too little coordination, and promising ideas never scale. Too much bureaucracy or monopoly power, and incumbents block disruptive innovation to protect their position. He applies this framework to the present, warning that both the United States and China may be drifting towards forms of sclerosis that could blunt the transformative potential of AI and green technologies.
INSIGHT AND ANALYSIS
At the core of the book is a simple but powerful idea: progress depends on the right match between institutions and the technologies of the moment. Frey shows how Britain’s relatively open markets and flexible political arrangements in the eighteenth and nineteenth centuries created space for entrepreneurs to challenge old orders. By contrast, imperial China’s powerful bureaucracy excelled at maintaining stability but ultimately stifled the very disruptions that might have renewed its economic dominance.
Translating that insight into today’s AI race, Frey is sceptical of the narrative that a more centralised, state-led model will automatically win. He acknowledges that strong states can be excellent at catching up and scaling existing technologies, but argues that genuinely new ideas tend to flourish in more decentralised environments with robust competition and tolerance for failure. When industrial policy hardens into protectionism and when corporate giants face too little challenge, innovation slows. In that sense, he sees current trends towards greater state control and corporate concentration in both the US and China as warning signs, not guarantees of an AI-driven golden age.
One of the strengths of How Progress Ends is the richness of its historical narrative. Frey moves confidently across centuries and continents, drawing on economic history, political science and technology studies. For readers who work in or around AI, this broader context is valuable. It reminds us that technologies do not exist in a vacuum. They interact with power structures, interests and social norms. A society can have world-class engineers and still squander an opportunity if its institutions reward rent-seeking more than experimentation.
The weaknesses of the book are, in some ways, the flip side of its strengths. Frey’s admiration for entrepreneurial dynamism can sometimes shade into an underestimation of the role that democratic restraint, social policy and environmental limits must play. Critics have argued that inequality, corporate capture and ecological overshoot are treated as secondary to institutional design, rather than as central drivers of stagnation in their own right. Even so, the core warning – that we should not assume AI will rescue us from our current problems – feels timely and necessary.
IMPLICATIONS
For policymakers, especially in emerging economies like South Africa, this book is a reminder that importing technology is not enough. The fate of nations depends on the quality of their institutions: their ability to channel new tools into broad-based prosperity rather than narrow advantage. That means competition policy, education, infrastructure, governance and trust matter at least as much as broadband, data centres and AI labs. Betting on a single “national champion” or copying another country’s industrial strategy without local adaptation is unlikely to work for long.
For business leaders, How Progress Ends is a challenge to complacency. It suggests that the current wave of AI enthusiasm could just as easily lead to a period of concentration and stagnation if incumbents use their advantages to entrench themselves rather than to open new frontiers. The lesson is uncomfortable: real progress often depends on being willing to disrupt one’s own business models, to invest in risky ideas and to welcome competition rather than seeking regulatory shelter.
For citizens, educators and parents, Frey’s argument cuts against the comforting assumption that each generation will automatically be better off than the last.
Progress, in his telling, is fragile. If we hollow out institutions, neglect public goods and tolerate extremes of inequality, technological miracles will not save us. That should prompt us to think carefully about the stories we tell young people about the future: not a guaranteed upward curve, but a field of choices in which AI is one powerful factor among many.
CLOSING TAKEAWAY
How Progress Ends is a big, demanding book, but it rewards the effort. Carl Benedikt Frey challenges one of the deepest assumptions in our AI conversation: that more computing power plus clever algorithms will inevitably produce a new era of abundance. His long view of history suggests something more sobering. Progress happens when societies balance experimentation with coordination, openness with stability, and when they allow new ideas to challenge entrenched interests.
It ends when institutions become rigid, when incumbents capture the rules of the game, and when we assume that the future will look after itself. For African readers, the message is clear. We cannot rely on someone else’s AI revolution to lift us. We have to build the institutions, skills and cultures that keep progress alive – or risk watching it stall.
Author Bio: Johan Steyn is a prominent AI thought leader, speaker, and author with a deep understanding of artificial intelligence’s impact on business and society. He is passionate about ethical AI development and its role in shaping a better future. Find out more about Johan’s work at https://www.aiforbusiness.net






Comments