Book Review: The Rise and Fall of the Artificial State — Jill Lepore on Government Without Consent

Executive summary

• The book: The Rise and Fall of the Artificial State, by Jill Lepore (Liveright, August 25, 2026; Allen Lane in the U.K.).
• The thesis: a system built on data, algorithms and concentrated corporate power has displaced parts of democratic governance, producing what Lepore calls "government without consent, even government without humans."
• The method: intellectual history, from the mechanical philosophy of the early modern period to today's AI ambitions, read through Hannah Arendt.
• Why it matters: it treats technology not as a neutral tool but as a political project with authors, interests and alternatives.

The argument

Lepore, the Kemper Professor of American History at Harvard and a staff writer at The New Yorker, opens with a scene borrowed from Hannah Arendt. In the prologue to The Human Condition, Arendt (1958) recalled the press reaction to the launch of Sputnik in 1957, which hailed it as a first step toward humanity's "escape from men's imprisonment to the earth." Arendt found that language alarming. Lepore uses it as the key to her book: a long tradition that sees the body, the place and the polity as limits to be engineered away.

According to the book and its early reviewers, Lepore traces that tradition to two ideas:

• Mechanical philosophy, the view that everything, including society, is a machine governed by quantifiable laws.
• Technological determinism, the belief that technology moves forward on its own and that progress follows automatically.

From these premises she builds the concept of the "artificial state": a form of power in which political campaigns behave like attention-mining algorithms, media platforms shape public discourse, and decisions migrate from accountable institutions to systems and owners that no one elected. Her central claim is historical rather than technical. None of this was inevitable, and what was built can be taken apart, as earlier systems of unaccountable power were.

The methodological value

The book's contribution is a change of question. Most debates about AI and politics ask what the technology does. Lepore asks who wanted it this way, and why.

That move connects her to a classic argument in the study of technology. Langdon Winner (1980) asked whether artifacts have politics, and answered that they can: the design of a bridge, a factory or a network can encode a distribution of power. Lepore extends that insight from objects to an entire mode of government. Algorithms are not just tools inside the state; they increasingly define what the state is able to see, measure and decide.

It also complements a book reviewed in this column last week. Matthew Botvinick's AI and Political Freedom looks forward, using game theory to anticipate where AI will put pressure on democratic institutions. Lepore looks backward, asking how the ideas that make such pressure possible were formed. Read together, they offer a mechanism and a genealogy.

Where the book is more demanding

The book has limits that readers should weigh:

• It is openly polemical. Publishers Weekly called it a "powerful anti-AI treatise." That clarity is a strength, but it means the case for technology's benefits receives less attention than the case against it.
• The literary sections are long. A substantial part of the book moves from history into myth and fiction, from ancient automata to The Hitchhiker's Guide to the Galaxy. Reviewers found it the most inspired part. Readers looking for policy analysis may find it a detour.
• The remedy is less developed than the diagnosis. Showing that a system was not inevitable is a necessary first step. It is not yet a plan for replacing it.

Practical implications for leadership

For governments, campaigns and organizations, the book suggests four questions worth asking now:

1. Where has consent been replaced by configuration? Which decisions that affect citizens, voters or customers are now made by default settings, ranking systems or automated rules that no one formally approved?
2. Who owns the instruments of perception? If your organization learns what the public thinks through platforms it does not control, its view of reality is borrowed.
3. Is "inevitable" doing political work? When a technology is presented as unstoppable, the argument often serves to remove a decision from debate. Leaders should ask who benefits from that framing.
4. What remains human by design? Identify the decisions that should never be delegated to an automated system, and say so publicly before a crisis forces the issue.

Verdict

The Rise and Fall of the Artificial State is a historian's intervention in a debate dominated by engineers and economists. Its value lies less in new data than in a reframing: the governance of technology is a question of legitimacy, not only of efficiency. It is recommended for anyone who leads public institutions, campaigns or communications teams, especially those who are tempted to treat the current trajectory of AI as a fact of nature.

References

Arendt, H. (1958). The human condition. University of Chicago Press.

Hughes, I. (2026, August 29). The Rise and Fall of the Artificial State by Jill Lepore: Courageous book offers visceral wake-up calls. The Irish Times. https://www.irishtimes.com/culture/books/review/2026/08/29/the-rise-and-fall-of-the-artificial-state-by-jill-lepore-courageous-book-offers-visceral-wake-up-calls/

Lepore, J. (2026). The rise and fall of the artificial state. Liveright.

Winner, L. (1980). Do artifacts have politics? Daedalus, 109(1), 121–136.