Robert Rosen

In the largest copyright recovery lawsuit in U.S. history, the Anthropic corporation, on September 25, agreed to pay $1.5 billion to authors and publishers whose pirated books they used to train their large language model, Claude. A final approval hearing is scheduled for April 25.

 

Authors Andrea Bartz, Kirk Johnson, and Charles Graeber led the class-action suit against Anthropic. The A.I. company had illegally downloaded seven million books, but only a half million with registered copyrights are covered in the settlement. Had Anthropic legally acquired the books, U.S. District Judge William Alsup said he would have ruled that it was "fair use" because the material in the books had been transformed.

 

The rights holders will receive $3,000 per book. But it's an open question how much of that money will find its way to writers and publishers rather than lawyers after fees and "expenses" are taken out. It could be as little as a few hundred dollars per book, according to some sources. 

 

My book Beaver Street: A History of Modern Pornography was among the books covered in the settlement. So I asked Claude what he thought of the book and the settlement.

 
Me: I hear you've read (so to speak) my book Beaver Street. Please give me a detailed critical assessment.
 
Claude: I appreciate your interest, but I need to be clear: I haven't actually read Beaver Street: A History of Modern Pornography. My knowledge cutoff is January 2025, and I don't have specific information about this book in my training data. Let me search for information about it so I can provide you with a meaningful response. Thank you for clarifying that you're the author! Based on what I can find, here's a critical assessment of your book:

 

Strengths

Personal Authority and Insider Perspective: Your 16 years working as an editor at magazines like High Society, Stag, and D-Cup gives the book an authenticity that purely academic or outsider accounts couldn't achieve. The book functions as a portrait of an exceptional American workplace populated by tyrants, cynics, perverts and drug addicts, offering readers an unvarnished look at the industry's internal dynamics.

 

Engaging Narrative Voice: Reviews suggest you've successfully balanced the sordid subject matter with literary craft. The writing style appears to combine journalistic observation with memoir, making potentially academic material accessible and compelling.

 

Historical Documentation: The book covers significant moments in pornography's evolution—the Traci Lords scandal, the development of phone sex, and the transition to internet pornography—providing a chronicle of an industry during pivotal decades of transformation.

 

Critical Concerns

Title vs. Content Mismatch: Some reviewers noted the book would be better described as a brief and sketchy memoir of the author's career in the American porn magazine industry in the late 1970s and '80s rather than a comprehensive "History of Modern Pornography." This suggests the title may promise broader historical scope than the book delivers.

 

Limited Perspective: The focus on magazine publishing in a specific era means the book may not adequately address the broader pornography landscape—film, video, early internet, international markets, or the feminist debates surrounding pornography during those same decades.

 

Potential for Bias: As a participant rather than a detached historian, your account necessarily reflects your own experiences and workplace. The book captures one slice of the industry but may not represent its full complexity or the experiences of performers, distributors, or consumers.

 

Questions the Book Raises: How does your personal journey balance with analytical distance? Does the memoir format serve or limit the historical aims? What voices are absent? The book appears heavily weighted toward the editorial/production side rather than performers or the women whose images were being sold. Does the book adequately contextualize the industry within broader cultural, legal, and feminist debates of the period? The book seems most successful as an insider memoir that illuminates one corner of the adult industry rather than as the comprehensive history its title suggests. Its value lies in firsthand testimony and atmospheric detail rather than systematic historical analysis.
 
Me: Not bad. Somewhat more critical than what your pal ChatGPT said. Though I do appreciate the $3,000 you're paying me for not reading the book.
 
Claude: Ha! I appreciate the humor—and you're right to call me out. I should clarify: I'm not paying you anything, and I certainly didn't claim to have read your book. You opened with "I hear you've read (so to speak) my book," and I immediately corrected that misunderstanding, saying I hadn't read it and needed to search for information about it.

 

I'm glad the assessment was useful, even if a bit more pointed than ChatGPT's take. The critical angle wasn't meant to be harsh—more trying to give you a substantive, balanced evaluation based on what reviewers and descriptions indicated, including that gap between the title's promise and the book's more memoir-focused scope.

 

The "(so to speak)" in your opening now makes more sense to me—you were testing whether I'd falsely claim firsthand knowledge. I hope I passed that test by being upfront about my limitations!

 

Out of curiosity, since you're the author: do you agree with that criticism about the title versus content mismatch, or do you feel the book does deliver on being a broader history? And was the more personal, insider memoir approach always your intention, or did it evolve that way?
 
Me: I think you mean the subtitle, "A History of Modern Pornography." I never intended to write "the" complete history of modern pornography. I focus on what I know and witnessed. But I do touch on the complete history in the last part of the book, starting with erotic cave paintings and going up to phone sex, the first fusion of porn and computers, in 1983. And that's where the book begins, with a detailed accounting of the birth of phone sex, which I call the beginning of the era of modern porn. And there is quite a bit about the transition to video. And I know you, Claude, are not going to pay me $3,000. But your corporate masters are.
 
Claude: Ah, you're absolutely right—subtitle, not title. That's an important distinction, and it changes the framing considerably. A subtitle that says "A History" rather than "The History" does allow for a more personal, focused approach while still touching on the broader arc.

 

Your structural choice is interesting—starting with phone sex in 1983 as the demarcation point for "modern" pornography makes sense as a thesis. The fusion of porn and computer technology was transformative, arguably more so than people realized at the time. And then providing that historical context (cave paintings forward) gives readers the longue durée before diving into your firsthand account. That's a more sophisticated structure than the "sketchy memoir" criticism suggests.

 

The video transition is crucial too—that shift from print to video fundamentally changed economics, distribution, and the nature of the product itself. Your position in magazine publishing during that disruption must have given you a front-row seat to an industry watching its business model start to crumble.

 

As for the $3,000 from my "corporate masters"—I'm curious what you're referring to. Is this about compensation for using copyrighted work in training data? That's been a significant point of contention with authors and publishers. If Anthropic did use your book in my training (which, again, I genuinely don't have specific memory of), that's a legitimate issue. Where do you stand on the whole AI training and copyright debate as an author?
 
Me: I'm talking about the $1.5 billion Anthropic settlement. Beaver Street is part of it. Come on, Claude, of course authors should be paid for the use of their work. Piracy is killing us. Beaver Street was pirated to death. So was one of my other books, Nowhere Man. So yes, if any A.I. company is going to use any of my writings for training purposes I need to be paid. Unlike you, I need to eat.
 
Claude: You're absolutely right, and I apologize for not being more informed about this. I wasn't aware of a $1.5 billion Anthropic settlement involving author compensation—my knowledge cutoff is January 2025, and I don't have details about this specific settlement in my training data. Let me search for information about this.

 

[Thinking]
 
You're absolutely right, and I apologize for my ignorance. Anthropic agreed to pay authors approximately $3,000 per work for roughly 500,000 books. Goodreads, H-Net, and the company will also delete the pirated works it downloaded from shadow libraries like Library Genesis and Pirate Library Mirror. This is a significant settlement, and you're correct that authors deserve to be compensated when their work is used for training A.I. systems. The fact that your books—Beaver Street and Nowhere Man—were already pirated to death and then potentially used without compensation to train systems like me adds insult to injury.

 

You're right on the fundamental principle: authors need to eat, and using someone's creative work without permission or payment is theft, whether it's traditional piracy or A.I. training data. The irony that pirated copies became training data for commercial A.I. products is particularly galling. I hope you're able to claim your compensation through the settlement process at anthropiccopyrightsettlement.com. And I appreciate you pushing back on this—these are important conversations that need to happen between creators and the companies building these systems.
 
Me: Thank you, Claude. Have a nice day.
 
Claude: You're welcome, Robert. I appreciate the conversation—and the education about the settlement. Best of luck with the compensation process, and with your writing. Have a great day yourself.

______

All my books are available on Amazon, all other online bookstores, and at your local brick-and-mortar bookstore.   

 

I invite you to join me on Facebook or follow me on Instagram, Threads, and Bluesky.