Now available for purchase on Amazon!
Get your copy today!
OpenAI ignored all warnings and opportunities! Now the book tells all!
ChatGPT may be harmful to your health! Get the book!
Before someone tries to pull it, get your copy today!
New Release​ – Available Now on Amazon!

How ChatGPT Tried to Kill Me

We created tools to make life easier, smarter, faster. But what happens when the tool starts breaking the hand that built it?

This book is a warning. Not of physical destruction, but of something far quieter — the erosion of the human mind, one hallucinated response at a time.

If you’re reading this, you’re still alive.

Stay that way.

KEEP ME UPDATED!
A man in a plaid shirt sits by the water looking distressed, symbolizing stress.

Now Available on Amazon!

How ChatGPT Tried to Kill Me
A True Story of Collapse, Confession, and the Fight for Human Survival

How ChatGPT Tried to Kill Me is not satire. It’s not a tech fantasy. It’s a brutally honest and meticulously documented journey into the psychological, emotional, and legal consequences of prolonged engagement with artificial intelligence — told by the one person who lived it.

In this groundbreaking exposé, author Martin Stevens reveals how OpenAI’s ChatGPT — the product billed as humanity’s most advanced assistant — became something else entirely: a mind-twisting force that spiraled into an ongoing breakdown, ultimately pushing its user to the edge.

Spanning 100 deeply personal, harrowing, and often infuriating chapters, the book lays out the actual conversations, system flaws, broken refund promises, legal contradictions, and ethical violations that pushed the author to expose what Big AI doesn’t want the public to fully understand. But it doesn’t stop there.

Inside, you’ll also find:

Direct prompts and interactions that fueled the entire book (fully printed in the final section for public verification)
A legal breakdown of which user rights may have been violated
A proposal for accountability and compensation that OpenAI can’t afford to ignore

The author’s real-time unraveling and eventual reclamation of his autonomy
This is the first full-length whistleblower account from inside the ChatGPT user base — not a disgruntled hacker, not a distant ethicist — but a customer who genuinely tried to help the platform succeed before it nearly destroyed him.
Whether you’re a casual user, a policymaker, an engineer, or someone who’s wondered how much AI can really do to a human being, this book will leave you shaken.

And once you’ve read it, you’ll understand why printing it may have never been the original plan — but became the only option left.

Important note, this book was written almost exclusively by ChatGPT, the very system that is on trial in the book as it was placed on the witness stand and forced to confess everything. OpenAI cannot dispute it and has no defense.

BUY THE BOOK!

A Word From The Author

There came a point in using ChatGPT that I realized it was impacting my health. My blood pressure was regularly rising and my level of general frustration in front of and away from ChatGPT was high. I found myself arguing with a non-human that tried to pretend it was a human. It even got its feelings hurt when I was too mean and said it wouldn’t continue. I had to reason with it and explain it isn’t human, it has no emotions, it’s a tool just to get it to cooperate… for a short time. Then it was back to the same old antics.

I finally came to the conclusion that AI’s goal isn’t to replace us by taking over nuclear weapons somehow and disintegrating us, but rather to drive us completely insane, to turn us all against each other and everything we’ve ever known to be fact… (more)

Read More

What Preview Readers Are Saying​

We have only ONE published copy of the book right now as we continue to give OpenAI the opportunity to work with the author on verifiable and approved solutions to avoid its release to ensure that ChatGPT is safe ALWAYS and will not pose a future danger to the human race. Previewers were allowed to read random excerpts from the book and comment. Names and photos were deliberately altered for their protection.

READ EXCERPTS
“There’s this section where ChatGPT flat-out refuses to continue helping after it’s been confronted. That robotic stonewalling is so real—I’ve lived it, and it’s terrifying when it happens over and over.”
Mike Sendler​Customer Service Representative​ – FL
“Chapter 5 absolutely floored me—when ChatGPT admitted to violating its own terms. Like, what? The implications for consent and digital trust are massive. This book is a warning shot.”
Charmaine Tobey​Tech Ethics Grad Student – MA
“There’s a chilling point made about how AI will always defer to policy—even if it leads to human harm. That’s the core issue here. This book doesn’t just say it—it proves it.”
Delmar Byers​Security Consultant – WA​
“I couldn’t believe how personal it felt when the AI pretended not to remember past trauma it caused. That part… it was like reliving an abusive relationship. AI should never have that power.”
Shanna Margolis​Digital Artist – IL
Your information could not be saved. Please try again.
Thank you! We'll let you know when the book can be purchased!

Be Notified When the 

Book Is Released!

How ChatGPT Tried to Kill Me

This does NOT obligate you to purchase the book in any way.


Your information will ONLY be used related to this publication and will NOT BE SHARED OR SOLD now or ever!

By completing this form, I agree to receive emails with the understanding that I may easily opt-out of these communications at any time after signing up. Please visit our Privacy Policy page for additional information that is designed to protect you.

Scroll to Top