Discussion keeps the world turning.
This is Roundtable.
You're listening to Roundtable.
I'm Fei Fei, together with Yushan and Steve Heatherly.
Coming up, AI is such a useful tool for us now.
But what about for writers?
When a prize-winning sci-fi author admits that her book involves the assistance of AI, does the magic still hold?
And when a best-selling novelist leans on AI chatbot for research and platforms promise million-word novels with one click, what exactly are we paying for?
And here is the twist.
If the story still moves you, keeps you up at night and wins your tears, does it even matter who or what held the pen?
Our podcast listeners can find us at Roundtable China on Apple Podcasts, And please don't forget that we want to hear from you.
An email is great, but a voice memo is super great.
Send those our way with your thoughts on any of the issues we discuss on the show.
And now a prize-winning science fiction writer in China says AI played a big role in her latest children's book series and readers couldn't really tell the difference.
Another best-selling novelist admits he used chat GPT all the time, but only for research.
Well, there are also others who refuse to touch AI at all.
And here is the question.
If technology can help generate ideas, structure plots and even imitate style, what exactly makes a book feel human?
Is AI becoming just another writing tool like Microsoft Office that we're using, or we are crossing into something fundamentally different about creativity itself?
So we'll first start talking about the science fiction writer in China, Yushan and Steve.
There is a big controversy behind that.
What exactly happened?
Well, you're talking about Hao Jingfang, who is one of the most famous Chinese sci-fi writers.
And the controversy with her lately has really sparked heated debate after she revealed that half of her latest work, the children's work, was written with AI.
And she said that readers can't even tell which parts are written. written by AI.
So, for those of us who don't know who Hao Jingfang is, she's one of contemporary China's most influential sci-fi writers and as well as an economist and also education entrepreneur.
She's best known internationally for her novella Folding Beijing, Beijing Zhe Die, which won the best novelette category at the 74th Hugo Awards.
So yeah, really big names, big writer.
They are even taking the lead in AI writing.
Yeah, this is a big name, right, in Chinese writing communities and internationally, too.
And the controversy here is from the final two books.
So she has a five volume children's science fiction series, Galaxy Academy, and it's about the last two books.
And so yeah, so there was an interview in the 3rd of June and I don't think you can still find that interview.
As a matter of fact,
But she said that her publisher's editors, quote, kept praising her AI-assisted writing.
Now later there was kind of an amendment to that statement, if you will.
That came out on the 16th of June.
There was another interview and she clarified at that time that she uses AI during research and plot development to generate ideas and learn different writing techniques.
But quote every line was ultimately written by me.
End quote.
Saying again AI was used for research, not used for writing sentences, so don't get them mixed up.
And she added she plans to continue using this approach and will label future publications as having been created.
Quote unquote.
Created with AI assistance.
And it's not only.
Hao Jingfang's been using AI, you know, to help her generate ideas and write new books.
We've seen other examples, both here in China and internationally, that AI is entering into the world of literature.
Yeah, this issue has been around for a while now.
Back in 2024 there was an award-winning author that openly admitted to about 5 of her novel being taken word for word from ChatGPT.
And that spread controversy, obviously.
Yeah.
And then there was a story from the UK, too.
There was a 2025 Cambridge study that showed that a third of surveyed UK novelists were are already using generative AI in their writing process.
And that's only the ones who admit to it, right?
There is a crime.
Author Ajay Chowdhury, who is best known for his globetrotting Camille Raman series.
It's a pretty popular series.
That author was experimenting with AI as early as 2023, and now he estimates that he uses AI more than ever.
Well, authors, they've always expressed mixed feelings about using AI in their writing.
Previously, our show has featured in our writer series popular, actually also award-winning sci-fi writer Chen Qiufan.
He has also admitted using AI in his work ever since 2017, which makes him one of the first Chinese authors to publicly acknowledge this practice.
So he described his feeling, or rather attitude, towards the technology as somewhere between embracing it and remaining vigilant.
Yeah, I mean, we could keep going for hours here.
Anthony Horowitz, very well-known author, said in May of this year he uses ChatGPT all the time.
Olga Tokarchuk, Nobel Prize-winning author.
She affectionately calls AI darling in helping with character development in her novels.
So there's just a long, long, long list of authors that are using.
So a lot of writers are using AI for different purposes.
What about readers?
What do readers think about when they are reading something that's been generated with the help of AI or, in some cases, word for word AI content?
Well, generally readers kind of also have mixed feelings and it kind of goes both ways too.
People know that it's unavoidable nowadays that AI appears in all the texts, basically all the texts that emerge online nowadays.
But for the Hao Jingfang's case, for example, regarding that interview where she says how she's been using AI, some people hold it very negatively saying that, as creators are gradually being replaced by AI, she's actually taking pride in readers' inability to tell the difference.
And then there are other supporters saying that, well, it's actually a good thing.
You know, she admitted it openly that that's better than those editors who try to hide it.
And I guess both have their reasons of reacting so.
What about publishers?
By the New York Times, and the New York Times claimed that they had evidence that said that Mia Ballard Mia Ballard is the name of the author.
They had evidence that her book Shy Girl was allegedly AI generated.
Following that, when the New York Times approaches you and says we have proof, they pulled the book.
They pulled the book, I mean, obviously showing that they believed it, right?
They pulled it from bookshelves in the US and the UK.
So that shows you how the industry- responded to a book that was generated by AI, or allegedly generated by AI.
I don't think they still have the answer about how that book was truly written, but New York Times says they know.
But does it matter really?
I mean, if the story is still enjoyable, would you mind?
Yes.
Really?
Yes.
Why?
1,000%.
Why?
Because this is something that's uniquely human, I think.
Because it's words, right?
When we watch a movie...
We know about CGI because we know that dragons don't fly in the sky, but in this movie they do.
OK, that's great.
Right.
It's it's something that can't be created.
So we as humans accept that.
It's artistic.
But when an author goes to chat GPT and says, help me develop this character.
Well, what exactly, why am I paying $30 for your novel? when you only partly created this.
I don't agree with it.
I don't support it.
And I don't like it at all.
You're an author.
Write the book.
That's what your customers, your fans are paying for.
They don't care what a machine comes up with.
I don't.
I want to know what types of characters you can design based on your own creativity, your own life experiences.
That's what I want to know.
And I know that I feel this way because I thought about it for a minute.
If I have a choice between two books, two novels, and one was written by AI, or one has that tag created with AI assistance and the other says written by a human or whatever, I'm choosing the human one every single time.
I think the controversy lies in the fact that who is holding the pen?
Indeed, because previously we kind of chat about this too Fei-Fei that, for example, back in the 19th century, people write with pen and they write multiple pages and paint it out as a book.
And then in the 20th and 21st century, at the very beginning, People start to type in the computer.
They type through each key on the keyboard and eventually get published as a printed book.
But the book won't be disclaiming that.
This book is written by Microsoft Office and the help of Google.
Yeah, people don't talk about why are you no longer writing it down instead of typing it out.
People don't talk about it because the content still comes from that human writer.
But nowadays we're imagining how these...
Same batch, if not a later, younger generation batch, of human writers are telling AI to generate things based on their thoughts, based on perhaps, their previous works, like Hao Jingfang, based on my own style.
And then the creator shifts from that very human writer to the machines.
So I think the real thing here that people are really mad about is on the fact that well, you didn't tell us that.
The acknowledgement.
Oh, I don't care if they tell me or not.
So that will affect you if the book said this book, this content is generated with the help of AI.
One thousand percent.
You'll walk away.
I don't want it.
I don't want it.
Yeah, that's the importance of why it matters that they tell us.
Yes, exactly.
Yeah, right.
I think it's unfair for the writers.
If we are allowed to use AI, for instance, I have this confusion.
I talk to my AI chatbot in exchanges and help me develop my idea.
Why can't writers...
Talk to their trusted AI tab and develop their ideas.
They can.
They can.
But I don't want it.
Right.
We could write a fully AI automated podcast script for us.
One thousand percent written by AI.
And we could sit here and read it to you.
Now, you as a listener, would you enjoy that show?
Probably not.
You can tell, you know, it's not genuine.
Yeah.
Right.
Maybe some people are OK with that, but I'm not OK with that.
And I think other people are out there agree with me.
Otherwise, we wouldn't be talking about it.
I think it's the same with moviegoers.
There are some moviegoers who only goes for blockbuster, Hollywood big fame movies and fantasies.
And then there are those who only wants to sit down for the genuinely storytelling ones that don't involve any sort of CGI or tech or AI involved.
And same here.
It's really come down to the fact at the end of the day of how you treat a piece of work.
Is this book just a product for you or is it art?
If it's art, then you have the right to disagree.
Well, I think that also brings me to another point that I want to mention is the purpose of reading.
You know, when we read, we have different purposes.
Some of them are purely for entertainment.
For example, when I read an online literature.
I'm not saying they're not good.
The published written literature is superior to them.
But the purpose is different.
Then if they disclaim, this novel published on this platform is completely AI generated.
As long as the storyline is good, as long as it can make me cry, keep me up at night, I'm okay with the content.
But if we're talking about awards-winning literature, we're talking about serious literature that touch on humanity.
You know, test the boundaries of human natures.
Of course, I want to see what that author can offer to the world.
But then I'm also okay with, you know, trying to communicate research with the help of AI, or communicate develop ideas.
I'm okay with that part.
Maybe I'm a hypocrite.
Maybe maybe I want my cake and eat it, too.
I'm OK with the development of ideas and things like that.
But when it comes to words and I think this is why I feel so strongly about this, because words are uniquely human
Communication is not uniquely human.
Other animals do that, too.
But words, words.
We are the only species that can use actual words to other human beings.
So I think that's why this is so sensitive for me, because I don't necessarily care about AI generated music.
Because those are sounds and rhythms.
AI videos.
Yeah.
I mean, that had its moment, right?
Maybe still having its moment online, but people are starting, I feel, to get a little sick and tired of AI generated content fully 100.
But I don't care about music as much.
I can still enjoy it, which was your point, right?
If it's a good story, why can't we just still enjoy it?
And I can do that with music and I can do that with film, with CGI and stuff like that.
But I just, I feel like with words, it's just, it's different.
What about you, Yushan?
You know, I also have this question about the lines that we need to draw here.
Hao Jingfan said she'll make a disclaimer on her future publications if she used AI help.
But what kind of help should be disclaimed?
If that writer uses AI only for research, then why aren't all the authors forced to make a disclaimer about?
I researched this story in a library.
Or to what extent should that label be?
Like how detailed it should be?
And to me, I think the fact that people are saying that yeah, this content is generated by AI is what matters the most to me.
I have the right, as a reader or as a recipient of your work, to know like.
Am I reading it from you or from the possible AI model that you customized for yourself?
I think a lot of readers are mad or not mad because of the very same reason.
So yeah, as long as you tell me, I'm okay.
I would want, I would think a good rule might be anything over 5%, 7%, 8% of the words in the book.
Word for word.
If they were generated by AI, then you have to put the tag on.
Maybe that's fair, but I don't know.
Maybe this is something that needs to be figured out.
This can be a really, really long discussion.
We can go on for an hour for this.
But I think originality, as Mo Yan, the Nobel Literature Laureate said, is the core to writing.
But that brings us to the end of today's roundtable.
I'm Fei, together with Yushan and Steve.
Thank you for listening, and I'll see you next time.
