The Human Warrant Series
Tested, Not Trusted
What Your Experience Is Worth in the Age of AI
A personal discipline for finding out which parts of your experience still hold up, now that AI makes knowledge cheap.
About the book
What your experience is still worth.
It starts the way these moments always start: a review room, a capable system, and an analysis produced in an afternoon that would have taken you a week at the same point in your own career. The work is good — not sloppy, not dismissible. And the first thing you do is look for the mistake, not because anything sounds wrong, but because it sounded too complete for someone who never had to earn it the slow way.
That quiet discomfort is the beginning of this book, not the end of it. Dr. Daryoush Tehranchi writes for everyone who has spent a career becoming the person people route the hard questions to, and who is starting to wonder — privately, honestly — what that is still worth now that the knowing part of it is available to anyone in seconds. This is not a book about whether AI will take your job. It is a book about the harder question underneath: what happens when what you know stops being the differentiator, and you have to find out — not assume, find out — what still is.
Tested, Not Trusted is a discipline, not an answer. Across thirteen chapters, it walks through the discomfort, the resentment nobody admits to, and the defensive instincts that follow — then turns toward something more useful than either false comfort or false alarm: learning to tell, in yourself, which parts of your experience are real, earned judgment that holds up under pressure, and which parts were only ever information that used to be scarce. Written across four decades in defense and space programs, it is the most personal of the Human Warrant books — and the one readers with real tenure keep saying feels uncomfortably like it was written about them.
Key takeaways
Six ideas worth keeping.
Takeaway 01
Some of what feels like earned judgment is just earned information.
Facts, patterns, and procedural knowledge that happened to be scarce for a long time can feel exactly like wisdom from the inside. The first distinction this book asks you to make is between inventory (what you know) and judgment (what you’ve actually tested against real outcomes) — because if you can’t tell them apart in yourself, you can’t tell them apart in anyone else either. (Chapter 1)
Takeaway 02
Name the resentment honestly — then interrogate it.
The feeling that surfaces when someone produces in an hour what took you years isn’t really about the work. It’s resentment that they didn’t have to earn it the way you did. The honest check: ask yourself, specifically, what you’d need to see to know whether this instance actually threatens something real. If you can’t name it, the resentment arrived faster than any actual answer could have. (Chapter 2)
Takeaway 03
Experience is the raw material, not the finished thing.
Years don’t automatically become good judgment — not for the thirty-year veteran, not for anyone. What converts time into something you can trust is the feedback loop: finding out what actually happened and letting it change you. What matters is whether the loop between expectation and outcome ever actually closed. (Chapter 5)
Takeaway 04
The confidence itself is never the evidence.
The same instinct, firing with the same conviction, can be exactly right once and exactly wrong years later. The only honest defense is a checked, current fact about whether the conditions that made the pattern true still hold. Slower than trusting your gut — and the only thing that actually distinguishes the first outcome from the second. (Chapter 9)
Takeaway 05
Feeling like you searched and having actually searched are not the same experience.
Even the best get fooled by their first idea: a room can spend forty minutes “considering alternatives” while unconsciously measuring every one against the failure mode spotted in the first five minutes. For decisions with real consequences, reserve the harder test — and consider asking someone who wasn’t in the room when the first idea took hold. (Chapter 7)
Takeaway 06
Asking to be challenged isn’t the same as being challengeable.
Running your decision memo past an AI assistant that returns five sharp objections can look like rigor while leaving the frame untouched — an answered objection is not an escaped one. And wanting challenge personally doesn’t help if disagreement gets filtered out three levels below you. It’s on you to notice when the objections reaching you have gotten too polite. (Chapter 11)
Who it’s for
For everyone who became the person people ask.
Questions
Before you ask.
Is this book about whether AI will take my job?
No — the author says so explicitly in the introduction. This is about something harder to see and easier to avoid: the moment when what you know stops being the differentiator, and you have to find out what still is.
Does it argue that experience doesn’t matter?
No. It refuses both easy stories — the comforting one that tenure is its own credential, and the alarming one that your years were wasted — as “easier books to write” that aren’t true often enough to be worth your time. The argument is a discipline for finding out, piece by piece.
Does it give a list of permanently “human-only” capabilities?
No — and deliberately. The book refuses that list on the grounds that every such list quietly gets shorter over time, and calls building your identity on one dishonest. What it offers instead is an ongoing practice of checking, not a verdict.
Are the scenarios in the book real events?
The book’s title page notes that the scenarios are illustrative composites — built to be true to the shape of real professional experience rather than accounts of specific individuals or events.
I’ve only been working ten years. Is this book for me?
It was written for people with real tenure — and Chapter 4 is aimed squarely at fast-rising younger engineers. The trap of getting better at looking good while the thing underneath stays where it was doesn’t require thirty years. For the companion argument about how expertise gets built, the same series’ Too Good, Too Soon continues it.
The author
Dr. Daryoush Tehranchi is an engineer, architect, educator, and author whose work focuses on complex systems, systems and software architecture, digital engineering, artificial intelligence, technical leadership, and the decisions that remain fundamentally human.
The Human Warrant Series
More from the series.
Connect with Daryoush