Future predictions for authors about publishing (AI writing, chatGPT legal issues and copyright)

Future predictions for authors about publishing (AI writing, chatGPT legal issues and copyright)

I don’t usually do these but this is going to be a big year of changes, for everybody in all industries. Here are some things I’ve talked about recently that will definitely change everything in 2023. If you’re a writer (or hope to be), you should be paying attention… things may be much easier and also much harder than they’ve ever been before, depending mostly on how you react to the marketplace changes.

PS. Don’t forget to check this post of big recent amazon changes.

How to publish a book (today)

Where are we? When I started publishing, things were harder and you had to do more work yourself, but it was OK if you didn’t do everything perfectly because there wasn’t as much competition. These days there are millions and millions of competitors. BUT – almost all books fail, including those from traditional publishing houses. You don’t need to compete with them (you just don’t want to be them).

But you’re still competing against other indie authors. Not really though, because we’re friendly and supportive and readers consume a ton of books. Just because other people are successful doesn’t mean you also can’t be successful. But you’ll absolutely have to have either better quality or more content, preferably both. Authors who just want to write a nice book and don’t care about sales, well, you can ignore just about everything. You don’t need information or advice. You do you.

But for authors who want to sell books, grow a fanbase, make a living, so they can write more books (because hey it’s fun and rewarding!) great – listen up.

  1. big publishers are barely ever taking on new authors, so you can spend years submitting, get a great publishing deal of 25K to 50k, which will be gone in three years and then start all over, or you can build your own platform and self-publish, if you avoid vanity presses.
  2. as always, the most successful indie authors find new ways to reach readers; write better books faster; get the best design they can afford, and some other stuff. It’s not as overwhelming as it all seems. If it seems too hard or impossible to market your books, there’s definitely a problem with your cover or blurb and those can be easy/cheap fixes. Indies can test and pivot faster, track results and keep trying till something works. But most authors won’t.

Everything everywhere all at once

I’ve been working on books and publishing tools for years and *most* things haven’t changed all that much in the past decade. But that’s not the case in 2023.

So what’s NEW this year? Well, everything, but specifically AI.

AI audiobooks narrations (text to voice)

Google has added AI voiced narration as an option to make audiobooks for awhile now if you publish with their platform. But the big news is Apple just launched the same thing for ibooks. And you may not know this but Amazon is the 3rd largest developer of natural text to speech voices.

Prediction: I think audible in 2023 will have an option to use AI voices for narration, for audiobooks on amazon. Why? because audiobooks are huge business but most authors can’t afford to get them made professionally. Amazon spent a ton of money launching their audiobook platform and are the industry leader right now, but that could all change quickly. With Google and now Apple on board with AI audiobook narration, Amazon will need to respond… and it already has the technology.

UPDATE: Amazon’s ACX is a rare platform that doesn’t allow authors to set their own prices; and they just made a big change by lowering prices. I would take this as a sign that they recognize their grip on the audiobook market is slipping and they’re attempting to stay competitive.

If you want to test out how good the voices are right now (keeping in mind they will be twice as good in six months) I set up a text to speech site. It needs a redesign and isn’t finished but the voices work: you can make a whole audiobook for about $30 (compared to $1000+). Is it as good as a real human? Probably not. Or maybe yes. That’s not really the question. The question is, is it good enough or will readers be pissed off? It’s probably a bit early, so you’re in the wild west. But in a year, the voices will be really good and this will be a thing.

I added a few samples: they aren’t perfect but they’re probably better than what you’re expecting. Don’t judge them harshly though. In a year, I doubt you’ll even notice they aren’t real humans.


Update: there’s already a new program from Eleven Labs that’s way better: you can clone anybody’s voice including your own and they sound so much more natural.

Ps. Audiobook narration from AI will probably grow quickly, but you could also put these up on your blog, on youtube, or wherever – just having the option to read your text out loud is a nice accessibility feature.

Ai writing tools & editing software

There are bigger discussions to be had about everything, and most artists are against AI everything on principle. But AI writing has gotten good and in 2023 it’ll become uncanny. You don’t have to agree with it, and you can totally hate it. BUT:

Bing just partnered with chatGPT3, so for the first time in 20 years we may have a google replacement, only that’s not quite true because chpt3 can do almost everything. Bing is owned by Microsoft, another ancient software that everybody hates but everybody still uses.

Prediction: I predict Microsoft will probably in-house something like prowritingaid or grammarly, but much better, and expand into text completion. Picture “clippy” – “it looks like you’re trying to write a book… want some help with that?”

Microsoft word may not get there fast. But in 2023 we’re going to see a bunch of super smart AI writing software that makes it easier than ever to research, brainstorm, outline, organize and yes – even generate – text so that you can put a book together faster. You don’t have to use AI… until you do. People are swearing they’ll NEVER use AI to write books, but what happens when it’s built into MS Word?

Canva just launched its own “Magic Write” tool, with an AI writer built-in. It works pretty well and I even checked it out through an “AI detection tool” and it came back as generated by human. There are also chatGPT3 plugins that work with Google docs; chrome plugins and WordPress plugins to generate blogs directly to your site.

Google, incidentally, already has versions of all this cool AI tech and they’ve been cautious with releasing it… but now they’re seeing competitors outpace and challenge them, and are rethinking their position. I wouldn’t be surprised if Google uses a version of chatGPT3 in search and google docs. Interestingly, Kodak invested the digital camera but didn’t release it because it would threaten their whole business. I think Google is in the same place now, with powerful AI tools that would make its whole business obsolete. So they’ve been slow to share, but are panicking right now.

The competition – and the public testing – will mean these tools get better fast, and there will be so many of them it’s hard to choose. AI writing tools to write whole books, are probably another year out, but they are already good enough to be useful for authors.

At the very least, I think within a year you’ll be able to proofread and copyedit your books easily with more advanced AI tools, that can just freshen up, fix and improve your rewriting, so you can either write a rough draft and tighten it up or generate a rough draft and manually improve it. You’ll still need some expertise to write a good book, but the annoying or challenging parts you avoid will be less daunting.

Obviously this is bad news for proofreaders and editors; but it’ll save authors a lot of time and money for authors.

PS. I’ve written over 20 books, about 2 million words, without AI. But like everyone, I deal with procrastination and frustration. I already use Vellum to format, even though I know how to format, because it’s smooth and easy. I use Grammarly, because I can’t spot all the typos myself: but I have to manually review and click hundreds of things and it takes hours. It’s stupid there isn’t an option to just upload your text and fix all the spelling, grammar and punctuation issues… but now there is.

I’ve used Google to brainstorm names, research places, find character and scene inspiration. Imagine a tool where all of this is built in, so you can waste less time? I’m not saying you should pump out AI written novels and publish them; right now a human editor is definitely needed to pull everything together and make it good. But in a year… I wouldn’t be surprised if it does a much better job of just continuing your story and even matching your own unique writing style and narrative voice.

Is it good writing? A lot of people are saying, AI writing is bad and basic; but like most AI tools, it depends how you prompt it. It’s capable of really great writing, if you have really long and detailed prompts. Most tools aren’t made for fiction writers (Sudowrite is one exception); but I’m building my own tool and the results are surprisingly good already. Plus you can do things like “summarize this chapter into an outline” or “give me developmental feedback on this to improve my writing.”

Keep in mind, the latest version of GPT3 launched a couple months ago and this is the first time it’s been not terrible; in six months I’m pretty sure it will be scary smart and very competent.

This article “AI is the end of writing” is pessimistic but fair: most commercial (successful/beloved) books that sell and are enjoyed, are written to formula and can absolutely be recreated and churned out with enough skill and experience. “Real artists/writers” who don’t write to market, don’t have much to worry about; the uncomfortable truth is that many writers are not writing good, clean, engaging content in any style, and that while chatGPT3 may be simple and boring, it is readable (and you can absolutely prompt it to do better.) It’s hard to even have this conversation because it’s easy to get lost in the theory, and much harder to ignore the results, but I can already generate samples like these…

*I felt like it was important to include, but you can scroll down and skip over the story to continue reading.

It was a summer night, and the stars shone brightly in the sky, but there was something sinister lurking in the shadows. The gates of the abandoned playground creaked in the breeze, eerily inviting any brave enough to enter. I had heard stories of this place, stories of a ghostly figure that haunted the playground at night, stories of strange creatures with a taste for human flesh. I had heard enough to know that I should turn back, but I also knew that if I did, I would never know what secrets lay beyond those gates. So, I swallowed my fear and stepped forward.

Little did I know that I was about to enter into a world of terror and mystery that I could never have imagined. The playground I had stumbled upon was deserted and eerily quiet. As I stepped closer, I could see why. The swing sets were covered in cobwebs, the merry-go-round was spinning without a rider, and the slide was slick with a thin layer of slime. It was as if a dark force had taken over the playground, and I was determined to find out what it was.

I heard a faint whisper on the wind, a whisper that seemed to be beckoning me. I stepped closer and the whisper grew louder. I could hear the words more clearly now, a strange and eerie song. Standing in the shadows was a figure shrouded in darkness, a figure with a strangely familiar face. I could feel the fear rising in my chest, and I took a step back. I opened my mouth to scream, but before I could, the figure stepped forward. The moonlight illuminated his face, and I realized who it was. It was the same figure from the stories, the ghostly figure that haunted the playground at night. He stared at me with an icy gaze, and I could feel my heart pounding in my chest.

I felt as if I was trapped in a nightmare, and I knew that I had to get away. I started to run, but the figure was too fast. He grabbed my arm and pulled me close. I could feel his breath on my neck, and I could smell the sweet scent of icecream and cockroaches. He looked at me with a strange intensity, and I could feel my body trembling in fear. I closed my eyes, expecting the worst, but instead, he whispered something in my ear. He said, “Welcome to my playground, you won’t be leaving here alive.” With that, he let go of my arm and stepped back into the shadows. I could feel the fear coursing through my veins, and I knew that I had to escape. I started to run, but I could still feel the figure’s eyes following me. I raced to the gates, and as I did, I could hear the laughter growing louder and louder. I made it to the gates just as the laughter reached its peak, and I raced away into the night, never looking back.


I’m NOT suggesting you write all your books with AI, but it can actually be great for very specific things: Here’s an article on how some Kindle Authors are using AI to write books. I was impressed by the prompt,

“I am writing a paranormal mystery that takes place in the small town of Table Rock, Texas. It has a female amateur sleuth. This is her name. I need a murder victim. I need how they were killed. I need four murder suspects with information about why they’re suspected and how they are cleared. And then tell me who the guilty killer is.” 

I tried that prompt and it gave me this:
The murder victim is a local rancher named Jacob Smith. He was found in his barn with a knife wound to the chest, indicating that he was stabbed to death. The four suspects are:

1. Lola Perez, a farmhand who had been working on the Smith farm for the past few months. She had a history of violent behavior and arguments with Jacob and was seen near the barn just before his body was discovered. She is cleared when it is revealed that she was out of town visiting family at the time of the murder.

2. Mike Johnson, a local businessman who was in a dispute with Jacob over a land deal. He was seen near the Smith farm the night of the murder and had a clear motive for wanting Jacob dead. He is cleared when it is revealed that he was at a meeting in the next town at the time of the murder.

3. Thomas Wilson, a former employee of Jacob’s who was fired for stealing from the Smith farm. He was seen arguing with Jacob the day before the murder and had a clear motive for wanting revenge. He is cleared when it is discovered that he has an alibi for the time of the murder.

4. Sarah Jones, a neighbor of Jacob’s who had been seen arguing with him the day before the murder. She had a history of jealously towards Jacob and had been heard threatening him. She is cleared when it is revealed that she was out of town visiting family at the time of the murder.

The guilty killer is Thomas Wilson, the former employee of Jacob’s who was fired for stealing from the Smith farm. He had a clear motive for wanting revenge and had been seen arguing with Jacob the day before the murder. He had returned to the Smith farm to confront Jacob and in a fit of rage, stabbed him to death.

That’s pretty useful, just to brainstorm the basic structure stuff, so you can focus on writing it yourself.

Is it plagiarism? There’s a moral argument to be made for sure. As for legal issues, I don’t expect Google, Amazon and Microsoft to spend billions of dollars and then lose a lawsuit. But also, it’s not currently breaking any copyright laws (style can’t be copyrighted + humans are allowed to look at things and be inspired). That said it’s not a good idea to use other (living) author names in your prompt. I’m building a tool to mimic famous classic writers, but mostly for educational purposes – you probably won’t have much luck writing as Mark Twain, and even if you do, none of the actual content is copied, only a vague, general style of that time period.

Incidentally, Mark Twain wrote a letter to Helen Keller after she was accused of plagiarism and said this:

“Oh, dear me, how unspeakably funny and owlishly idiotic and grotesque was that “plagiarism” farce! As if there was much of anything in any human utterance, oral or written, except plagiarism! The kernel, the soul—let us go farther and say the substance, the bulk, the actual and valuable material of all human utterances in plagiarism. For substantially all ideas are second hand, consciously or unconsciously drawn from a million outside sources and daily use by the garnerer with a pride and satisfaction born of the superstition that he originated them; whereas there is not a rag of originality about them any where except the little discoloration they get from his mental and moral calibre and his temperament, which is revealed in characteristics of phrasing.”

Many writers already use some form of the hero’s journey, which Joseph Campbell developed after noting almost all powerful mythologies and stories share a common structure. I can’t find the specific quote I put in my PhD thesis, but a few centuries ago some smart guy said something like “want to write a great story but don’t know where to start? Just take any classic myth and reframe it with new characters in a modern setting.”

Should new laws be written? Sure, but if they are, it would impact the entire internet and all writers and artists would be in trouble for creating anything (since our output is always conditioned on all the media we’ve consumed).

There’s also Roald Dahl’s short story, the Great Automatic Grammatizator, about a story-writing machine, first published in 1953. This stuff used to be fiction and fantasy, but now (suddenly!) the robots are proving capable of creating art and writing. You should read it, because it’s very timely, but here’s the dystopian ending:

And worse is yet to come. Today, as the secret spreads, many more are hurrying to tie up with Mr Knipe. And all the time the screw turns tighter for those who hesitate to sign their names. This very moment, as I sit here listening to the howling of my nine starving children in the other room, I can feel my own hand creeping closer and closer to that golden contract that lies over on the other side of the desk. Give us strength, Oh Lord, to let our children starve.

Basically, when AI writing is a thing – like it is now – it will be so successful that every living writer must either get on board or starve. It’s a stark and dire warning against this kind of technology. But not because the writing is bad. Because it’s too good.

PS. Here’s how you could actually write a whole book with chatgpt3, with some writing samples.

AI art (text to image)

AI art is more unpopular and contentious, because it’s visual. And while most artists say AI art has no soul and can easily be discovered, the truth is I keep seeing AI art in all kinds of groups (architecture, cats, cake, room decoration) and for most people it just looks like amazing stuff: it gets tons of comments, shares and likes, because it looks great and it’s something new and better than what most people have seen before.

The most fascinating thing for me is this real comic from 1923, accurately predicting the exact year (2023) that AI art will be an actual thing.

Gizmodo wrote an article about this comic in 2014. At the time, it didn’t seem like something that would become possible in the near future, and yet it has arrived right on time…

Like many futuristic cartoons from the early 20th century, this one is more spoof than sincere — if anything a commentary on the inherent weirdness of outsourcing creativity to machines. But joke or not, I guess we’ll have to wait 9 years until Webster’s prediction can officially be tossed on the failed futures pile. Sometimes the most outlandish predictions have a way of coming true.

Just so you have an idea of the quality of images being produced by AI art and text to image generation, here are some examples I’ve found online; not all of these are Midjourney; some are stable diffusion, which is an open source software anybody can install on their desktop. Some are creative; some look like photographs. While controversial, it’s important to recognize that the big stock photo companies, like Shutterstock, are mostly all already on board, either accepting AI art or having their own in-house tools. Canva and Microsoft’s new “Designer” tool have text to image generation. Meta, Nvidia and several others are launching their own versions.

Getty images is the exception, having raised a lawsuit, but they have their own history of theft and plagiarism, and I believe it’s mostly a money grab to cash in on the trend or try and preserve their images to build their own unique tool.

The images above are made with midjourney AI, here’s a list of the prompts I used. And here’s my post about AI art for book cover design.

You can avoid making AI art yourself, but it will be tough to avoid in the real world, for all kinds of designers, and why buy it from a stock site when you can generate something unique yourself? Using it commercially is currently a gray area, as it’s legal according to current copyright laws but many artists are expecting sweeping changes to protect human creators. Even so, we’re already seeing authors use AI character art and stuff for kickstarter campaigns, promotions, etc. This will become way more common and start being used on book covers as well, even by traditional publishers. Right now, the legal stuff is murky and most people are skeptical (or rightfully outraged).

PS. Already if you buy a cover on Fiverr or 99Designs, there’s a good chance you’ll get AI generated art in submissions and it’ll be very tricky to know whether you’re getting “real” art and design…

You can’t copyright AI generated art, but in 2023 it’ll become very difficult to avoid, even if you resist. The 1923 comic about an art-robot and “idea dynamo” showed the artist going fishing, because he was able to finish his work faster, but Roald Dahl had it right: this is a huge threat to all artists and photographers. For the past few decades, creatives could either try to find clients for custom work, or sell digital assets as stock. Even if a lawsuit is settled favorably for artists – in terms of paying them a tiny percentage each time their name or art style is used – they will be signing themselves out of a job.

Who benefits: the big businesses, but that’s who has always benefited; and they “win” by providing services that everybody else enjoys using enough to pay for.

For most authors, this means you’ll be able to (maybe) get better and cheaper book design art, illustrations and other cool stuff than you ever could have before. You’ll also be able to make promo graphics easier and better than ever. But, given the controversy, it’s probably best to use it for personal writing inspiration only, or perhaps to share character art based on your book. There are ethical ways to use AI (not using specific artist names) but you may get negative comments about AI in general, even if you’re not using it commercially.

Prediction: the metaverse is kind of a joke right now, but I expect some version of it to succeed in the next few years or so, and this is the real allure of AI art. Imagine being in a virtual world where you can conjure up any images, like a dog riding a bicycle or a human riding a shrimp? It would be impossible for companies to predict everything you might ask for, but AI generation could do it – Google is already working on a text-to-video software and they aren’t the only ones. YES, this is a huge threat to all creative industries, content creators, illustrators, and all kinds of artists; it’s as equally dangerous for all writers. No doubt this is a shift like we’ve never experienced in our lifetimes. It’ll be a rocky year, but when the dust settles, more humans will be able to express their creativity; though it also means they won’t need to pay experts to help achieve their vision.

Imagine if you could say “tell me a bedtime story about X” and it would generate the text, images, and then read it out loud to you? We’re very close to that being a reality right now. Or “create a powerpoint presentation or sales pitch to close a client” and it would do all of that.

We are in a simulation…

I recently watched the “White Noise” movie with Adam Driver, based on the book by Don DeLillo, which I wrote a paper about in grad school. There’s a lot of French Postmodernism, ala Jean Baudrillard.

Simulcra are copies that depict things that either had no original, or that no longer have an original. Simulation is the imitation of the operation of a real-world process or system over time

One poignant quote from the movie is this:

“Don’t we deserve attention for our suffering? Isn’t fear news?”

Another fascinating thing is the tendency to identify AI art by counting the fingers and toes; just like we were taught to do with Fairies and the supernatural. Or to try and read text, like we would do in a lucid dream to tell if we are asleep.

We are in the uncanny valley, but now the rules of reality are shifting; the monsters are here with us.

I’ll point out several worrying examples:

  • CNET was accused of publishing AI articles for months, I’m sure they aren’t alone. Interestingly, there’s a newer post accusing them of plagiarism and claiming the AI was simply a data-scraping respinner of content. I would argue that the new GPT3 is much more sophisticated than what they were using; and that all of the examples show standard facts/advice that are already on millions of sites across the internet. How do define something like “overdraft protection” without giving a simple, standard response; do we all need to be extra creative with everything so we never repeat words? Who “owns” the copyright for common knowledge like that?*
  • Apparently, Buzzfeed has also announced they will be using AI to write articles; but I first saw a fake news report generated by gpt3 on Reddit and then the “real” news media picked it up; and I can’t tell how much of it is real.
  • An Artist Spent 100 Hours Working On A Book Cover Image, Only To Be Accused Of Using AI
  • “AI Detectors” give false results, (it’s very possible to create AI content to pass them or for human content to be marked as positive). Some of them are outright fakes.
  • There are AI lawyers (why wouldn’t there be; if it’s about knowing facts and writing a persuasive argument). But it will get very meta if the AI robot lawyers argue for or against their own existence.
  • A *real* copyright case, Dr. Dre forces Taylor Greene not to use his music. This is how the law should work, when someone is using your content directly for their own gains, often without attribution.
  • But it gets really messy, when you consider that 90% of the art on the internet is fan-art; pictures of copyrighted cartoon and movie characters. If a case rules AI art is illegal, it will give Disney, Marvel etc the power to rebuke or sue every creator of fan art; what is currently protected as fair use could block all art completely (but I don’t think that will happen).
  • Microsoft says their AI can simulate any voice based on a 3 second audio. Very realistic deep fakes are coming very soon; but there will copyright and legal issues around them as well (it’ll be impossible for famous people to block everything made from their image, though they could go after the platforms)

I expect we will see a lot of litigation and news about this stuff over the next year, but not to ban AI tools outright; only so dying businesses can claim a piece of the pie before they are replaced. The interesting thing is this: can regular humans tell the difference? If it’s legal for humans to copy, clone (even “collage” is currently protected, where real images are used, and AI art is far beyond that) – if it’s legal for humans but not robots – first we’ll have to prove an image or some writing was written or created by AI, and that’s going to become very difficult.

I believe the main thing will be, do people feel wronged, abused, or cheated: is the creation close and similar enough to a famous person, artist or writer that fans can recognize it and are duped into thinking it’s original and sanctioned? Did they not get the value or quality they expected?

I’ve also seen some ads for oil paintings that look like Midjourney, and it’s probably the case that some working artists are adapting the AI style to be trendy; I expect we’ll see AI in art galleries just like we saw pricy NFT’s go for thousands of dollars last year – because art has almost always been what’s new, on the cusp of creative technologies, and what people respond to emotionally. The AI controversy *almost* proves on a meta-level that it is real art, because it pisses people off and makes them uncomfortable. That’s probably a weak argument, but I do think in a few decades this will be remembered as the time of “AI art” as an artistic style.

* artists and writers have always claimed that it’s their suffering, their process, that gives their work value – the amount of time and struggle it took them to achieve mastery; but in practice, people don’t care how hard the work was to create, they only judge the final output – and consumers will often choose the best, most affordable option, even if they feel a little bad about it.

Is AI writing plagiarism?

I linked to the above article on CNET’s AI plagiarism, but there’s a lot more to unpack there, and it’s a complex issue. The article quotes a professor who says the content “clearly” rose to the level of plagiarism, and that a student at a university would be sent the ethics committee or kicked out. But there’s a difference between what might be considered unethical plagiarism that would get you in trouble at school and actual plagiarism, as a possible criminal act. CNET probably won’t get into any legal trouble.

But that’s not the main thing, it’s this: imagine AI that can scrape through your books and search the entire internet to find matches. Of course it will find some, even if only by chance. There’s only so many words you can use. Imagine having to respond to it, and rewriting your own writing to make sure it never used a phrase that’s ever been used before; avoiding the simple and practical in favor of the strange and creative? For weird fiction, a fun experiment. But for an article, opinion piece or blog? Paralyzing.

Here was an example:

“Can You Buy a Gift Card With a Credit Card?”

vs

“Can You Buy Gift Cards With a Credit Card?”

Should we never try to answer basic, common questions because it’s been covered before?

Or:

What is overdraft protection?

“Overdraft protection is an optional feature offered by banks to prevent the rejection of a charge on a checking account with insufficient funds.”

vs

“Overdraft protection is an optional service that prevents the rejection of charges to a bank account… that are in excess of the available funds in the account.”

If this is plagiarism, can we not count on any factual common definitions but always invent some new answer, even if false? We’d have to delete the whole internet…

So if you wrote: “It was dark. The sun had set hours ago. A drop of rain wet my cheek and I looked up to see storm clouds brewing.” You might get in trouble because “It was a dark and stormy night…” already existed? Copyright laws already protect “transformative” work which is a very low bar, basically changed at all. But more importantly, readers are used to certain phrasing that go unnoticed. It used to be the case, a century ago, for writers to cram in as many uncited references and phrases as possible, to show off their extensive knowledge (and make readers feel good when they got the reference).

I’m not saying plagiarism is OK; it never is, at all. But with the rise of super computers, I think it’s dangerous to focus on this issue or claim all AI writing is plagiarism because it was inspired from somewhere else; just as our own writing inevitably has been. (The same danger for artists, who would ban AI but might accidentally prevent themselves from making any fan art from comics, video games or movies, which is very common, if all art must be 100% inspired from nothing.)

It’s a valid fear and concern, because our brains can’t do what computers can do, so we don’t understand where the content is coming from and it *could* have a sentence lifted from somewhere else – though that’s highly unlikely – and even in this case, a similar phrasing for a sentence out of a whole book could never get you into legal trouble; which leaves the ethical argument. And that’s entirely fair: to say you won’t use AI because it was inspired by human creativity without credit or compensation. That’s a personal, humanist argument that is valid, for YOU. But it doesn’t have a legal basis, so it’s unlikely to affect actual policy or implementation.

2023 publishing AI


The year of the water rabbit

It’s supposed to be a year of peace, prosperity and hope. Honestly, I don’t see that right now. There’s too much turmoil and transition. Most of the big tech firms are firing thousands of people and replacing them with automation, both robots and AI.

ChatGPT3 is almost good enough to act as a constant companion, like a genie in a bottle with infinite resources and wisdom, that can actually do stuff for you. When it’s paired with AI voices, we’ll have smart devices that actually work and sound human. It’ll probably replace call centers. It’s already being developed into therapy apps, though as one of my writer friends points out: “THERAPY? WHAT. Like, all of the minds behind AI have all said that it is in its “teenage” years and still learning. How in the world can anyone trust AI to give therapy and not respond with something horrific? We’re just going to be like sure, let’s literally put vulnerable people’s well-being in the hands of what is basically a super smart, apathetic teenage sociopath?”

And there’s this, another friend posted a list of 10 chatGPT3 quotes. Most of them were fine, simple puns or dad jokes. But #9 was “how do you make a plumber cry? Kill his family.”

Even if it gets answers right 9/10 times, that one wrong answer can be brutally out of place. ChatGPT3 is already much less effective than it was at first, because they keep add limitations so people can’t access sensitive information or disinformation (though there are always workarounds). Hopefully, it’ll get smart about contextualizing answers, depending on prompts, but right now it can’t really be trusted.

And if you account for Boston Dynamics new demo robot, we’re getting very close to a dystopian/futuristic “I-robot” scenario.

Interestingly, since that demonstration is used on a mock-construction site, I’ve seen discussions from builders saying things like “a robot could never replace a real builder because they need to make decisions on the fly and react to unexpected events; make complicated decisions, etc.” These conversations are happening in every industry, but the artists/writers feel it more than most because creative expression is part of our core identity; the thing that makes us human and ourselves, the thing that gives us joy, not just a job.

Google has changed its stance on AI writing and said they don’t mind it, but they focus on value, which I think is how AI everything will go.

If it’s valuable to the reader, then it’s OK. This has always been my approach to writing and publishing as well: if readers are happy; if people like and enjoy it; if they’re happy with the experience – that’s the target goal you need to shoot for. It’s true there will be a flood of basic crap to Amazon and Google, but I would argue it’s not worse that the many poorly written articles and books already out there.

The bar will be raised, which means you’ll need to produce better content than your competitors, which has always been the case. Last year, Tiktok was the hot new platform to sell books; now some people are talking about banning it. What will it be this year? I don’t talk much about book marketing anymore because I’m still obsessed with productivity (I can’t sell until I finish).

I’m not sure how AI will affect my businesses yet but I do know the way I do things – the hard, slow, manual way – is exhausting, and I’d be open for making it more fun or easier, especially if it reduces avoidance, depression or tedious, uncreative, banal work that I hate doing.

The interesting thing, for me, is the contrast between art as therapy and art as business: people say they only use art to express themselves and the only value of Real Art is because it gives purpose and meaning to humans. Those people, and that art, or that feeling at least, cannot be replaced by AI, but it can be stolen by AI (the fact that robots can do it faster and better is demoralizing, and therefore psychologically harmful). But that’s a different argument than, “they’re taking our jobs!” – that implies art as produce, art done on demand for a specific client’s request, which isn’t necessarily creative though it has always taken a high amount of skill. Working artists have always struggled with the balance.

Nobody is stopping you from continuing to persist and enjoy what you love (or perhaps, they are: I recognize that AI everything is an existential threat to human identity). I don’t have answers, but my opinions matter very little. I side with artists and writers; but I’m also against gatekeeping, misinformation and outright bullying. These tools may have taken you by surprise, and I sympathize with whatever emotional responses you’re feeling. But at this point, these new tools will probably impact you and your business specifically, regardless of how you feel about them. And they could be powerful tools to help you create your best work, but you’re free not to use them if you prefer not to.

I’m not defending AI art or writing, but I am inherently opposed to blanket statements of animosity (or outright violence) that paint all creatives who experiment with new tech as AI bro’s, soulless demons, thieves and hacks – as one redditor put it, “if you use AI for literally any purpose in your writing you are morally and artistically bankrupt.” A similar Facebook comment reads “No writer with an ounce of morality or self respect would use AI to write their book, not even as a first draft.”

But even then… I get it. AI is being opposed on ethical grounds, and the battle might be waged in terms of moral philosophy – and it’s a sharp edge. However I fear the ones who might be hurt the most are those with the most to gain.

As Stanford University’s Erik Brynjolfsson says in this well-rounded CBS report,

“I think we’re going to have potentially the best decade of flourishing of creativity that we’ve ever had because a whole bunch of – people lots more people than before – are going to be able to contribute to our collective art and science.”

Paraphrasing, he also likes to use the word “affected” not “replaced” because it’s not really about AI replacing artists and writers; it’s going to be about artists and writers who use AI replacing those who don’t. But also: there’s a good chance midjourney is not for artists, but people who are not artists; and GPT3 is not for writers and authors, but people with the aspirations who lack the necessary skill or talent.

It is terrifying for people who have developed valuable skillsets to be replaced so easily; just as it is absolutely marvelous for people without the skills or talent to suddenly see a huge boost to the quality of their creative content. It can’t help but be a controversial issue, and I doubt the controversy will go away easily. But since, arguably, there are more people without skills and talent that those with skills and talent, this means AI is a product with a vast and valuable use-case, that people will be eager and willing to pay for. Some countries are better at protecting their citizens, but most are driven by commerce and capitalism.

I probably feel as jaded and cynical as Roald Dahl, “Give us strength, Oh Lord, to let our children starve” – but that’s really the reality you’re facing if you oppose AI tech at this point: are you willing to sacrifice income and success by refusing to use very lucrative tools that improve your life (to experience hardship; not just for you but those you care about?) I used to be, absolutely, as a starving artist: my identity as an artist and making what I was passionate about (even when nobody would buy it) was stronger than my wish to live in safety and comfort.

I made a personal choice – and realization – that real art is what people value (or you can convince them to value) and that real creativity should be well paid (and if it isn’t, you probably aren’t making anything remarkable, regardless of your core beliefs). That’s not nice to hear, so most authors prefer the romantic ideology that is single-handedly responsible for the era of starving artists. One of my favorite articles on AI writing puts it this way:


“Our exaggerated reverence for the creative impulse derives from the romantics of the early 19th century… AI should disillusion us of the spurious glamour of creativity… I also hope it helps us to see that being creative does not inevitably imply that you are unusually interesting or unusually intelligent.”

JAMES MARRIOTT, the Times

Conclusions….

To summarize, in 2023 publishing is going to be changed by new tech that makes it much easier and much cheaper to publish. The biggest costs to publishing have always been:

  • editing
  • cover design
  • audiobooks
  • formatting

It’s always been possible to “hack” these or DIY, but professional publishing services will cost you an average of $2000, and unscrupulous vanity presses with encourage book marketing packages for twice that. New tools on the horizon will make it easier to plot, edit and even rewrite your own work without an expensive developmental editor. Sure, maybe it’s not nearly as good: but for all the authors who have no budget for editing, it’ll allow them to publish better quality, cleaner manuscripts faster. It won’t threaten people who do have the budget to do things the “right” way and there will probably always be bespoke, boutique, “human” services. But when the robots become better than human services – which they may – it will be hard to resist getting the best, cheapest service in favor of a less-good, more expensive service “with soul.”

Your opinions and beliefs about AI will probably reflect whether you’re someone who is benefited by, or harmed by, these new developments. In a CNN/Anderson Cooper interview about chatGPT3, the guest said something like “what needs to happen, and what we’ve never really done well, is to find a way to support or retrain those who are displaced by new technologies,” and I think that’s a great sentiment.

We should be careful, both in the sense of emotionally sensitive, and also by approaching AI tools with a healthy dose of caution and skepticism.

PS. Right now I’m just in the discovery/fascination stage, these new tools are too new and there’s too many of them to really feel comfortable putting them into my writing or workflow process. But if you do want to learn exactly which tools are best and how to use them to write and publish books, I’m adding a section about AI into my course, the 21 Day Author Platform.

UPDATE: free AI writing prompts generator…

I built a tool for educational/entertainment purposes. It’s a writing prompts generator but you can edit the prompts and use AI to write anything. Hope you like it!

AI writing prompts generator

Add Comment

Your email address will not be published. Required fields are marked *