AI is pumping out books. Are they any good?

Summary

Podcast Summary: AI Is Pumping Out Books. Are They Any Good?

Podcast: The Indicator from Planet Money
Date: June 25, 2026
Hosts: Waylon Wong, Adrienne Ma
Special Guests: Imka Reimers (Cornell University), Joel Waldfogel (University of Minnesota), Jeremy Tarr (Fodor’s Travel)
Total Length (excluding ads): Approx. 9 minutes

---

Episode Overview

This episode explores the impact of large language models (LLMs) on the book publishing industry, focusing on the dramatic increase in the volume of ebooks—particularly those attributed to AI—and examines whether these AI-generated books are actually any good. The hosts approach the topic from an economic standpoint, featuring insights from academic researchers and a human editor at a travel guide publisher.

---

Key Discussion Points & Insights

1. AI vs. Human Writer Prolificacy (00:06 - 00:44)

• Hosts compare famously prolific human authors (Stephen King, Isaac Asimov, Barbara Cartland) to the high-volume potential of LLMs, pointing out that models can produce output on a scale that far surpasses even the most productive human writers.
• Statistic: New book releases on Amazon nearly tripled between 2022 and 2025, a surge attributed to LLMs.

2. Researching the E-Book Boom (02:06 - 04:32)

• Researchers: Imka Reimers and Joel Waldfogel, experts in the digitization of cultural products, discuss their experiment in self-publishing.
  • Imka Reimers (on her own self-published book):
    “It was a half page about my day. It was nothing.” (02:26)
  • Finding: After the launch of ChatGPT, Amazon’s monthly new ebook releases increased from approximately 100,000 to 300,000 by mid-2025.
    • Joel Waldfogel: “And then essentially in lockstep with the increase in the use of ChatGPT, we see it increasing by late or mid-2025 to 300,000 new books released per month.” (03:39)
• Methodology: They used AI detection software to estimate that the growth in new ebooks matched the proportion identified as AI-written, leading to the conclusion that AI was the main driver behind the surge.
  • Imka Reimers: “Almost one for one. It was really just something.” (04:17)

3. Assessing Quality: What Does “Good” Mean? (04:32 - 05:47)

• Economic Definition: The researchers consider a book “good” if it appeals to some people, i.e., if some people buy it, regardless of literary merit.
  • Joel Waldfogel:
    “What we tend to mean by good when we talk about this stuff is appealing to some people... So it’s not the same as the good that a cultural critic would say...” (04:50)
• Results:
  • AI-generated books had:
    • Lower star ratings
    • Fewer ratings overall
    • Worse sales ranks compared to human-written books
  • Joel Waldfogel:
    “By all those measures, the AI books are less useful or in some sense worse than the non-AI books.” (05:40)
• Conclusion: AI books, on average, are not “good,” and human authors are not being displaced (yet).

4. No Breakout AI Hits—Yet (05:59 - 06:43)

• Observation: Unlike previous digital disruptions (YouTube for music, digital fan fiction for books), AI-generated books haven’t produced viral successes or hits; instead, they cluster in mediocrity (“middling stuff”).
  • Joel Waldfogel:
    “AI doesn’t seem to be like that. It’s giving us more products, but they’re not products all the way across the distribution. It’s a whole bunch more near the middle.” (06:14)
• Caveat: The field is evolving fast—future AI may produce higher-quality, more differentiated content.

5. Perspective from a Human Editor: The Travel Guide Example (06:43 - 09:15)

• Guest: Jeremy Tarr, Digital Editorial Director at Fodor’s Travel
  • Sees a notable influx of AI-generated travel guidebooks, many suspiciously prolific (authors with 100+ guidebooks in a short time).
  • Jeremy Tarr:
    “Knowing how much work and effort goes into producing a guidebook... it’s impossible.” (07:23)
• AI as Competition: Tarr isn’t threatened by AI travel guides but notices travelers use tools like ChatGPT for trip planning.
• New Solution: Fodor’s is launching their own AI chatbot, “Eugene,” trained on Fodor’s proprietary content, to provide more reliable, curated, and opinionated travel guidance.
  • Jeremy Tarr:
    “I think human-backed AI is the best kind of AI you can get... So AI helps speed that up, it helps organize it. It’s just a new way of finding the information.” (08:17) “Eugene is friendly, but no nonsense. I think a lot of AI is fairly sycophantic and we wanted to make an AI that is opinionated.” (08:55)

6. Memorable Moments and Host Banter

• Humorous banter:
  • Waylon Wong (re: Fodor’s AI): “I can’t wait to have my travel plans totally crushed by Eugene the chatbot.” (09:15)
  • Waylon Wong (jokingly channeling the chatbot):
    “That’s a terrible idea. Only a dumb person would go to this museum…” (09:21)

---

Notable Quotes (with Timestamps)

• Imka Reimers [02:26]:
  “It was a half page about my day. It was nothing.”

• Joel Waldfogel [03:39]:
  “And then essentially in lockstep with the increase in the use of ChatGPT, we see it increasing by late or mid-2025 to 300,000 new books released per month.”

• Imka Reimers [04:17]:
  “Almost one for one. It was really just something.”

• Joel Waldfogel [04:50]:
  “What we tend to mean by good... is appealing to some people... 50 Shades of Grey, by this definition... is undoubtedly a very good product because lots of people bought it.”

• Joel Waldfogel [05:40]:
  “By all those measures, the AI books are less useful or in some sense worse than the non-AI books.”

• Joel Waldfogel [06:14]:
  “AI doesn’t seem to be like that. It’s giving us more products, but they’re not products all the way across the distribution. It’s a whole bunch more near the middle.”

• Jeremy Tarr [07:23]:
  “Knowing how much work and effort goes into producing a guidebook... it’s impossible.”

• Jeremy Tarr [08:17]:
  “I think human-backed AI is the best kind of AI you can get... It helps organize it. It’s just a new way of finding the information.”

• Jeremy Tarr [08:55]:
  “Eugene is friendly, but no nonsense... we wanted to make an AI that is opinionated.”

---

Key Timestamps for Important Segments

• 00:29: LLMs’ impact on book publishing volume
• 03:08: Methodology for detecting AI-written books
• 04:32: Assessing “quality” of AI books
• 05:23: Data: AI books have lower ratings, sales
• 06:14: Distribution: No breakout AI success stories
• 07:18: Publisher perspective on the glut of AI travel guides
• 08:17: Launching a human-curated AI travel chatbot
• 08:55: Designing Eugene’s “personality”

---

Summary Takeaways

• The massive increase in e-books is driven almost entirely by AI, not humans.
• By both economic and consumer measures, AI-generated books are currently lesser in quality (sales and ratings) than human-authored counterparts.
• There have been no standout AI literary hits; the flood of content is largely middling.
• Publishers like Fodor’s are adapting by combining trusted, human-driven content with AI tools to offer curated, authoritative experiences—countering the impersonal sludge of mass-produced AI books.
• The conversation remains open-ended: as AI evolves, so might the quality—and commercial viability—of its literary output.

Transcript

[Imka Reimers]

Npr.

[Waylon Wong]

Think about some of the Western world's most famously prolific writers. You have horror novelist Stephen King. He's written more than 60 bestselling books and hundreds of short stories.

[Adrienne Ma]

There's also sci fi author Isaac Asimov, who produced more than 500 fiction and nonfiction books, and also English romance novelist Barbara Cartland, who wrote more than 700 books during her career.

[Waylon Wong]

So much swooning. But today, large language models or LLMs, can match that kind of output in a single year. The number of new book releases on Amazon almost tripled between 2022 and 2025. That's an increase that can be attributed to LLMs.

[Adrienne Ma]

But are these AI written books any good? This is the indicator from Planet Money. I'm Adrienne Ma.

[Waylon Wong]

And I'm Waylon Wong. Today on the show we tackle this quality question through an economic lens. Plus we hear from a human editor at a niche publisher how they're fighting the slot wars.

[NPR Announcer]

This message comes from Edward Jones where they believe rich means opening yourself to new possibilities. That's why your dedicated financial advisor meets you where you are, helping you move forward with confidence. Let's find your rich Edward Jones Member

[Advertisement Voice]

SIPC this message comes from LinkedIn. Owning a small business comes with a lot of challenges and means juggling multiple things at once. It's even harder to do it efficiently, but with LinkedIn you get all the tools you need to grow in one place. With LinkedIn, simplify your sales, marketing and hiring so that you can actually run your small business. Learn [e-mail] indicatorshow this message comes from Dell Technologies. Interruptions happen at work, but with the Dell Pro laptop powered by Intel Core Ultra, with vpro, built with optimized battery and built in intelligence, you your tech won't slow you down. Dell.com Dell Pro built for you Economists

[Waylon Wong]

Imke Reimers and Joel Walvogel have worked together for more than 15 years. Imka's at Cornell University and Joel's at the University of Minnesota. Together they've studied how cultural products like music and books have gone digital. And sometimes this has involved some firsthand experience. As Imca explains, I tried out to

[Imka Reimers]

see how easy it is to self publish a book, not to try to sell any units. To be very clear to just say oh yeah, I self published a book. Please don't buy it. What was your book about? Oh it was a half page about my day. It was nothing.

[Waylon Wong]

I feel like you're underselling it, but okay.

[Imka Reimers]

I really am not.

[Joel Walvogel]

Didn't you get two five star reviews?

[Imka Reimers]

Yes because two people did not listen to my plea not to buy it.

[Adrienne Ma]

Well, it sounds like IMCA is part of this wave of self publishing. Because the number of self published books skyrocketed in 2008 following the launch of Amazon's Kindle e Reader. Then 2022 brought another big disruption. The widespread availability of large language models like ChatGPT.

[Waylon Wong]

Imga and Joel wanted to see if this new technology was also leading to lots more books being published. And if there was an explosion in AI generated books, were they any good?

[Adrienne Ma]

We'll come back to the issue of good in a bit. But first, IMCA and Joel needed to figure out whether LLMs were driving a big increase in ebooks. They focused on Amazon, which has two thirds of the ebook market. And they figured out the number of new ebooks going on sale each month. Before the release of ChatGPT, it was about 100,000 books per month.

[Joel Walvogel]

And then essentially lockstep with the increase in the use of ChatGPT, we see it increasing by late or mid-2025 to 300,000 new books released per month.

[Adrienne Ma]

That is a tripling in the number of ebooks for sale.

[Joel Walvogel]

This includes all sorts of books, including those published in the traditional ways, as long as they have an ebook manifestation. And of course, everyone always releases an ebook. But most of these books are self published works published through Amazon.

[Waylon Wong]

The next step was determining how many of these books were written by AI. So Joel and Emka ran a sample of around 50,000 titles through AI detection software.

[Imka Reimers]

What was really interesting there is that the share of books determined to be written by AI really matched the growth rate that we saw in the new books being created. Almost one for one. It was really just something.

[Adrienne Ma]

In other words, this boom in new ebooks could be attributed to AI. So now this brings us to the meaty question in Joel and Imka's research. How good are these books?

[Waylon Wong]

And the word good here can mean so many different things. Joel, explain how an economist like him interprets it.

[Joel Walvogel]

What we tend to mean by good when we talk about this stuff is appealing to some people. Some people would buy it. So it's not the same as the good that a cultural critic would say, but, you know, 50 shades of gray, by this definition, the economic definition is undoubtedly a very good product because lots of people bought it.

[Waylon Wong]

I'm going to go pick up a copy of 50 Shades of Grey and see if you gave a blurb for the back that says, like this book is good.

[Imka Reimers]

Economists like this book.

[Joel Walvogel]

Exactly. No, economists understand that people like this book. Yes, that's, I think, the blurb we would give it. This is a valuable product.

[Adrienne Ma]

Imka and Joel used star ratings and public Amazon sales rankings to measure whether people were buying these new AI books. And the results were clear. AI books had fewer ratings than human written ones. The average star rating for AI books was worse and their sales rank was lower.

[Joel Walvogel]

By all those measures, the AI books are less useful or in some sense worse than the non AI books.

[Adrienne Ma]

In other words, they were not good. So Joel and Imka say their research suggests that human authors don't have to worry, at least for now, about being displaced by AI.

[Waylon Wong]

Another interesting thing is there doesn't seem to be breakout hits emerging from the AI slush pile. Jewel points out that, you know, Justin Bieber first got discovered on YouTube and 50 Shades of Grey. Remember, a good book had its origins as Twilight fan fiction published on the Internet.

[Joel Walvogel]

These are artists who made it through, came out of left field, as it were, and were made possible by digitization. AI doesn't seem to be like that. It's giving us more products, but they're not products all the way across the distribution. It's a whole bunch more near the middle. But, but having said that, everything is pretty new. So, you know, AI changes every day. And so maybe future AI will be more like regular digitization and less like big piles of middling stuff.

[Adrienne Ma]

Even these piles of middling stuff, though, have caught the attention of people like Jeremy Tarr. He's the digital editorial director at Fotors Travel.

[Waylon Wong]

What was your last trip?

[Jeremy Tarr]

My last trip was a road trip from Lake Como to the French Riviera.

[Waylon Wong]

That sounds so glamorous.

[Jeremy Tarr]

It was very glamorous.

[Waylon Wong]

No need to rub it in, Jeremy. So Fotors has been publishing travel guides for more than eight decades, and travel is one of the categories where there's been an especially big increase in AI generated ebooks, according to Imka and Joel's research.

[Adrienne Ma]

Jeremy at Fotor says he's skeptical of many of the new travel guidebooks he's seen online.

[Jeremy Tarr]

If you were to see just one or two of them, it'd be, you know, somebody's just entrepreneurial and they want to make a guidebook, which is, that's great. But we. You then look at the author and the author has published, you know, 100 of them just all over the world in a very short period of time. And knowing how much work and effort goes into producing a guidebook, whether it's, you know, 160 pages or 500 pages, it's impossible.

[Waylon Wong]

Jeremy says he doesn't see AI travel books as a threat, but he does know that people are relying on ChatGPT and similar tools to plan their trips. So in July, Fotorz is launching a new AI product called Eugene, named after late company founder Eugene Fodor. It's a chatbot trained on Fodor's guidebooks and digital content and it will give travel recommendations and help help make itineraries.

[Jeremy Tarr]

I think human backed AI is the best kind of AI you can get. It's like reading an article, reading the guidebook. I mean, there is so much information on any given city and it can take a really long time to plan something. So AI helps speed that up, it helps organize it. It's just a new way of finding the information.

[Adrienne Ma]

Jeremy says some of the large language models out there have already scraped Fodor's material without the company's permission. Building a proprietary AI tool means the company can make money from this new technology and it can control everything from the information it feeds Eugene to how the chatbot sounds.

[Jeremy Tarr]

Eugene is is friendly, but no nonsense. I think a lot of AI is fairly sycophantic and we wanted to make an AI that is opinionated. If you say should I go here or here? They will give you an honest assessment.

[Adrienne Ma]

I can't wait to have my travel plans totally crushed by Eugene the chatbot.

[Waylon Wong]

That's a terrible idea. Only a dumb person would go to this museum. Okay, you say so Eugene.

[Advertisement Voice]

Whatever you say.

[Waylon Wong]

This episode was produced by Julia Richie with engineering by Kwesi Lee. It's fact checked by Sierra Juarez. Keegan Cannon is the show's editor, and the indicator is a production of npr.

[NPR Announcer]

This message comes from Capella University. That spark you feel? That's your drive for more Capella University's flexpath Learning format lets you earn your degree at your pace without putting life on pause. Learn [e-mail]. this message comes from St. Martin's Press, publisher of how to Try Again by Steve Kami. Some people wake up at 4am, run 15 miles and optimize every moment of their day. How to Try Again is a book for everyone else. It's about quitting unhealthy expectations, failing compassionately, and trying again differently. Blending empathy, humor and real advice, how to Try Again will help you make change that sticks. How to Try Again by Steve Kam is available wherever books are sold.