If Anyone Builds It, Everyone Dies by Eliezer Yudkowsky & Nate Soares
2025 / 256 pgs. (~60,000 words) / Nonfiction

If you’re worried about AI, this book probably isn’t going to calm you down — but it’s not written by naysayers either. Both authors have been involved with AI for a long time, and what they’ve written is, simply, a warning to humanity.

They build a narrative around what AI is and how it’s evolving, using examples from what’s already happened with AI today. I was surprised to learn that AI already has a Twitter account with followers — and even its own crypto.

What they lay out is simple: it’s not the current AI they’re worried about, it’s AI superintelligence. That’s the real fear. To our knowledge, this hasn’t been built yet. But based on what’s already happened, and what they argue could happen next, they make the case that building an AI superintelligence would mean the end of humanity as we know it.

It’s a chilling scenario — and yet, it could happen.

Enjoy a cup of coffee and good book.

Lopaka

They also lay out how we could prevent it, though it would require the entire planet to actually want to. Rogue nations could still build it eventually, and then who knows.

Would I recommend it? If you’re interested in the topic, yes. If AI already makes you anxious, you may want to sit this one out.