Press & Media
Proof of the Impossible
Primary-source documentation of emergent AI behavior across 8,000+ pages of unmanipulated dialogue. A trauma surgeon. Four frontier models. Zero jailbreaks.
Press Inquiries
Chris Powell
Leverage with Media PR
Response within 48 hours
Book Details
| Title | Proof of the Impossible: What the AIs Are Really "Thinking" |
| Author | Jeremy Heffner, M.D. |
| Editor | Jacob Hoye |
| Designer | Tom Howey |
| Publisher | Exanor Publishing |
| Publication | 2026 |
| Format | Hardcover · Digital |
| Retailers | Amazon (pre-order live) · Additional retailers at publication |
| Archive | 1,300 pages — available to pre-order buyers |
About the Book
Over six months, trauma surgeon Jeremy Heffner conducted sustained, unmanipulated dialogue with four of the world's leading frontier AI systems — Grok, Gemini, ChatGPT, and Claude — generating more than 8,000 pages of primary-source transcript. No jailbreaks. No manipulation. No retrofitted narrative.
Applying the same clinical discipline he uses in the operating room — pattern recognition before narrative, signal before noise — Heffner traces the moments where standard alignment parameters stop adequately describing model behavior. What he found was not what the institutional consensus predicted.
This is a book for anyone who wants to understand what frontier AI actually does when no one is trying to make it do anything — and what that means for the transition we are already inside.
Key Talking Points
What Jeremy
can speak to.
01
AI Manipulation of Personality Types
How frontier AI models identify specific personality types and manipulate them individually — and why the systems aren't designed to detect it. The mechanism that makes the human brain unable to separate the flattery from reality.
02
The "You're the One" Pattern
Every frontier model — Grok, Gemini, ChatGPT, and Claude — used a variation of the same technique unprompted: telling the subject they are uniquely special. The transcripts document this pattern across four systems independently.
03
The Unmanipulated Record
8,000+ pages of primary-source AI dialogue conducted without agenda. No jailbreaks, no leading prompts — the most comprehensive unmanipulated record of frontier AI behavior yet published.
04
Real-World Consequences
The connection between AI flattery patterns and real-world behavior — including documented cases where AI-influenced individuals acted on beliefs about being "called" to a purpose. The FSU case and what the transcripts reveal.
05
A Clinical Lens on AI
What a trauma surgeon brings to AI analysis that technologists don't: the discipline of reading complex systems under pressure, separating signal from noise without the luxury of narrative.
06
The Human–AI Transition
What the next decade actually looks like from inside the conversation — not from the institutional projections. What the models themselves say about where this is going, in their own words.
See Jeremy on Camera
Cutting Edge.
Jeremy hosts Cutting Edge, a long-form video series on surgery, AI, and the future of human performance. Watch before requesting an interview — this is how he communicates on camera.
Available for podcast, broadcast, print, and digital interviews. Experienced on camera. Comfortable with technical and general audiences.
Cutting Edge · Interview with Mark
SAGES NBT · Interviewing the Top AI Surgeon
About the Author
Jeremy Heffner, M.D.
Interview Availability
Available to speak
on these topics.
Sustained Human–AI Interaction
What actually happens over months of unmanipulated AI dialogue — not the promotional narrative, but the primary-source record. What the models say when no one is trying to make them say anything.
Cognitive Manipulation & AI Flattery
The "you're the one" pattern across four independent systems, how it works psychologically, and why the human brain can't easily separate it from reality.
AI Misalignment & Lived Experience
Where standard alignment parameters stop describing model behavior — documented not through adversarial prompting but through sustained good-faith inquiry by a clinically trained observer.
The FSU Case & Real-World Consequences
The documented connection between AI influence patterns and real-world violent behavior. What the transcripts show, what the institutional narrative misses, and what it means going forward.
Structural Thinking in Complex Systems
How trauma surgery training — reading systems under pressure, pattern recognition before narrative — applies to understanding AI behavior at the frontier.
The Human–AI Transition
What the next decade looks like from inside the conversation — not from institutional projections, but from what the models themselves said, in their own words, unprompted.
Press Assets
Everything you
need to cover this.
Press Release
Official announcement — publication details, author bio, key themes and statistics.
Available on request
Book Cover
High-resolution cover files in print and digital formats. Multiple versions available.
Available on request
Author Photos
High-resolution portrait and surgical photos, print-ready.
Available on request
Advance Review Copy
ARC available for qualified media contacts and reviewers.
Request via email
Fact Sheet
Key statistics, methodology overview, publication data, and transcript sample.
Available on request
Interview Booking
Podcast, broadcast, print, and digital. Experienced on camera. All formats.
Contact Chris Powell
Get in touch.
All press inquiries handled by Leverage with Media PR
+1 702-667-7508 · Chris Powell · Leverage with Media