Afraid of the Right Things

On a Tuesday evening in October, four people read the same headline about machines that help build machines. Renee thought of her job. Isaiah thought of the first job he still did not have. Joanne thought of her daughter upstairs, who some evenings talks to a chatbot. Walt thought of the films in which nobody could find the switch.

None of them could say how large the thing they feared actually was.

Fear of AI sounds like one fear. It is at least six, traveling under one word. Fear for your work, for your own ability to think, for the truth, for your personal information, for the people you love, and fear of losing control of the technology itself. They rest on different evidence and call for different responses. Carried around as one large, shapeless dread, they are hard to act on.

What this book is

This book takes your fear seriously. It explains in plain language what is actually happening. How a system like this is made, what it can and cannot do, and what it means when a company says its AI is helping to build the next one. Then it takes each fear in turn and says what is known, what is disputed and what nobody knows yet. You do not need to know anything about computers to follow it.

What it gives you

• five questions to ask of any headline about AI, worked through with real headlines from the past year
• what the numbers on jobs show so far, and where the first signs of change seem to be
• what a family can do about cloned voices and fakes, starting with a family word
• what chatbots keep from your conversations, and the settings worth checking once a year
• how to talk with your children about chatbots, starting with asking them to show you the app
• the people most worried about losing control and their critics, quoted in their own words, and the strongest form of each case
• a simple grid for sorting your own fears by how well they are supported and how much you can do about them

It also covers what the tests of these systems have and have not shown, why experts' estimates are so far apart, who holds the money and the power, how much electricity and water all of this uses, and who is writing the rules. It describes the state of things as of October 2026, gives a date and a source for its evidence, and says where it is offering its own judgment and where no study has tested its suggestions.

What it is not

It is not a warning that the end is coming, and not a promise that AI is harmless. It does not ask you to be less afraid. It asks you to be afraid of the right things, in the right measure, and to know which of them you can do something about.

A vague dread gives you nothing to do. A smaller, sharper fear is something you can live with and act on.