What if AI doesn’t decide what is true — but decides which version of truth becomes impossible to ignore?
The Consensus Machine explores how artificial intelligence absorbs the information structures around us — repetition, institutional authority, ranking, citation, and consensus — and turns them into answers that can sound far more certain than the evidence underneath them.
This is not a book about whether AI lies. It is about something more subtle: what happens when machines inherit our information hierarchies and begin reproducing them at enormous scale.
Artificial intelligence doesn't have to manipulate us to reinforce what we already believe. It only has to inherit an information system where repetition, authority, and visibility often shape what we accept as truth.
In The Consensus Machine, we explore how AI can amplify dominant narratives, marginalize less visible perspectives, and turn repeated claims into seemingly reliable answers.
A thought-provoking look at how to distinguish genuine evidence from manufactured consensus—and use AI without surrendering our own judgment.
Go beyond the pages.
Your digital copy includes interactive resources designed to help you put the book into practice. Work through them at your own pace, save your responses, and take your experience further.
See the machine at work
The book opens with a direct confrontation between a user and an AI system over a contested geopolitical event.
The important question is not simply whether the AI's answer is correct. The important question is why the system sounds so certain, which sources receive the most weight, and what happens when the user begins challenging the dominant narrative.
From there, the book follows the information backward — through media repetition, authority signals, ranking systems, citation cascades, training data, and feedback loops — to reveal how apparent consensus is constructed.
