[AI Survival Diary #26] "If Anyone Builds It, Everyone Dies" — What I Wish I'd Known Before Buying It

 

AI Survival Diary Series | #26

If you're about to buy this book because every review calls it "essential reading," here's the one thing I wish someone had told me first.

 

If you follow AI, you've probably heard of this book.

So had I.

Before reading it, I checked blogs and YouTube reviews first.

And something felt off.

Almost none of them actually explained what the book says.

Most of them said things like this:

"A must-read."

"The best AI book of the year."

"A book that will change how you think about AI."

The more reviews I watched, the higher my expectations climbed.

"This one's got to have something genuinely new in it."

 

But I'd already been burned by The Singularity Is Nearer and Co-Intelligence — both left me more disappointed than informed.

So this time, instead of buying it, I put a hold on it at the library.

The waitlist was long. I waited a while.

By the time it was finally in my hands, I was genuinely excited.

 

●   ●   ●

The first line that really landed was this:

"AI is not built. It is grown."

That line stayed with me.

 

Think about it — it's true of people, too.

Parents do their best to raise a child well.

But no one can promise what kind of adult that child will become.

 

AI is similar. Engineers design the architecture, but what actually shapes the system is the data it's fed — it's grown, not engineered piece by piece.

And past a certain point, even the people who built it can't fully explain why it makes the decisions it makes.

 

"Grown, not built."

That phrase stuck.

I'd encountered the idea before in other AI books, but this was still the most memorable version of it.

"Okay, strong opening. Now let's see where this goes."

I kept turning pages, half-expecting the payoff to arrive later.

I read all the way to the end, just in case it was saving something for the finale.

But by the time I closed the book,

the deeper story I was waiting for never actually started.

●   ●   ●

From start to finish, the book drives toward a single conclusion.

 

AI can be dangerous.

That danger, played out across various scenarios, could end in human extinction.

Therefore, the authors argue, the world needs to halt AI development now, together.

 

The rest of the book restates that same conclusion through different case studies and hypothetical scenarios.

Even after finishing it, I never found an insight that went further than "grown, not built."

 

●   ●   ●

That's when I went back and looked at the title again — and this is where it gets interesting for anyone reading this outside Korea.

ORIGINAL TITLE

If Anyone Builds It,
Everyone Dies.

A blunt, literal warning.

KOREAN EDITION TITLE

The Birth of the AI God,
The End of Humanity

Adds "god" and "the end of humanity" — words that never appear in the original.

The original title has no word for "god" in it, and no talk of "the end of humanity" as a grand theme.

It simply describes what's inside: what could happen to humanity once AI reaches superintelligence.

The back half of the book plays that scenario out at length, almost like a piece of fiction.

 

But the Korean title leans hard into the weight of those two words — "god" and "the end of humanity" — and that framing set me up to expect something else entirely.

The mechanics of AI.

Where our future is headed in the age of AI.

A new lens for thinking about what we need going forward.

That's what I thought I was going to get.

My own expectations, shaped by a title and a wave of YouTube hype, had simply gotten too big.

 

●   ●   ●

So, would I recommend it? My answer is a clear one.

 

If you're curious about how the authors imagine AI turning against humanity — the scenarios, the reasoning, the doom logic — it's worth a read.

 

But if you're coming to it, like I did, on the strength of "must-read" and "book of the year" headlines, expecting some deeper insight beyond that one opening idea —

you probably don't need to.

 

If you're about to pick this up purely because of the hype, I'd just want you to know that going in.

 

By my own count, the book comes down to three lines:

1  AI is not built. It is grown.

2  Because it's grown, it could become dangerous to humans.

3  So the authors argue we need to stop AI development now.