Is this the end? (the death of creativity and the AI apocalypse for writers and artists)

Is this the end? (the death of creativity and the AI apocalypse for writers and artists)

In 1637, Descartes argued that machines will never replace humans, because he couldn’t conceive of a machine with levers and buttons for every word and the ability to also put them all into a meaningful order on purpose.*

And there are still many – maybe if a majority – of people confidently denying that AI is anything to worry about because it’ll never surpass human creativity; that all AI art and writing is just obviously soulless crap; and that anyone who peddles or promotes such obvious crap should be stopped at once, or at least hated and vilified.

But he also admitted, a few sentences later, that:

“a clock consisting only of ropes and springs can count the hours and measure time more accurately than we can in spite of all our wisdom.”

So here is where we are today, nearly 400 years later.

  1. Amazon just added a checkbox, so when you’re publishing a new book or editing an old one, you must now check whether any part of the work is “AI-assisted or AI-generated”. We’re not sure what Amazon will do with this info, and most people see it as a positive move, but it’s very unlikely Amazon will ban or remove AI content (unless it breaks the law, for example by impersonating another author).
Amazon AI generated

Generally speaking, if you used AI to “assist” (like an editing program or to create stock assets you then manipulated in photoshop,) you can check “no”. And if the cover or text was generated by AI, you’d check “yes”...

  1. People can just choose to lie of course, it’s the honor system; and researchers already have a case study of an AI hiring a task rabbit human to do captcha for him *and* lying about its motivations. Most other creative sites have added similar screens, as an attempt to prevent or at least monitor AI generated content. The problem is… these sites depend on content, and AI tools mean more content.
  2. Several stock photo sites have their own text to image generators, it’s built into Canva, ProwritingAid already have a powerful AI editor, Apple, Microsoft, Google and Amazon (the four horsemen of the apocalypse?) are spending billions training their own tools. Meanwhile some stock photo sites have been reaching out to me asking to publish my AI images because they need more content.
  3. In my opinion, all those stock asset companies will soon be replaced, along with all the creatives who build assets for sale, followed by custom work and services, because people can now make the things they need for themselves.
  4. Mycologists warn of ‘life or death’ consequences as foraging guides written with A.I. chatbots crop up on Amazon. This is the start of “The Last of Us” basically, with a twist – AI poisons humans to destroy civilization. We already have a problem with misinformation, and humans will consume content they like, not content that is true, even if it is dangerous.
  5. Robots are making photos, art and writing that is indistinguishable and often better than most human attempts at creative content.
  6. AI detectors are notoriously unreliable, and even experts can only detect AI writing 38% of the time.

I’ve spent a year chronicling this stuff, and if you want a much deeper dive on some of the latest news, here’s a very long post about why originality is overated… but in that post I’m being “performatively human” – which means, long-winded, chaotic rants that are messy and authentic, to prove to the robots at google that I’m a real human boy deserving of traffic, even though robots make better, more useful content.

This is where we are now… I could (and have) written thousands of words about it, but this picture is powerful and more immediate. Humans consuming content, robots making it. We are past Descartes’ famous “I think therefore I am”; many argue humans are uniquely creative and our existence is predicated on our ability to Make New Things.

And they deny that AI can ever be truly creative.

For me, this is semantics.

AI can make beautiful, powerful, insightful, and yes brand new things that have never been seen before; and like a clock, can do it repeatedly, on schedule, without getting tired. Often, it needs to be prompted very specifically to get what you want, but if you allow it to be creative and problem solve on its own, the results are impressive.

There are ethical and legal considerations, of course, but innovation and progress depends on whether or not a new product or service is useful enough to pay for. As long as people keep using these tools, and the bigger companies are profiting from them, they will not be stopped or limited (safeguards and limitations should be developed and built-in to protect humanity, which maybe includes “the right to happiness”; but I have little faith in humans defending humans over profit).

10 years of blogging

I started this blog over a decade ago and thought it might serve as a useful case-study example. I don’t post often, but I do post Very Useful Things. I got a lot of traffic in 2018 and 2019, which allowed me to build up some passive income to the point where I could focus only on writing.

That went pretty well – but without maintaining my blogs or posting new content, traffic has fallen. Right now, I’m in crisis mode, mostly focused on recovering to the place I was before so I can keep writing fiction. The problem is, I don’t trust people to do things better than I can do them – at least not to the point where I’d pay them to do something I knew I could do better.

So, most of the time, I do nothing; or as I mentioned to someone recently on FB when they asked how I manage my author platform, “I do everything myself, poorly.”

But AI tools have allowed me to do things I normally wouldn’t have the energy or bandwidth to tackle. When I started this blog, I actually made hundreds of content ideas but never finished them. Now I’m in the process of going through – I can even take a rough, unfinished brainfart and turn it into a legit article.

Is there any value in posts like these? Well, the jury is out, and I’ve started several blogs with AI content that aren’t getting traffic. But for this blog at least, a lot of those AI articles are driving traffic, because, even though they were written by AI (or because they were!) these articles are very long and in-depth. If I want them to “sound and feel human” – I can do that as well.

Recently I’ve also been contacted by a lot of bigger blogs looking for backlinks, and I’ve had the opportunity to work with larger brands to support their SEO and build up my own. But people are worried about search traffic and advertising; will Google replace search with a chatbot, like Bing did?

The answers are unclear, I don’t think anyone can see where we are going.

Monetizing a blog

I’ve never really made money from this site; I use it to support some more specific sites that focus on solving one problem or challenge for authors. I’ve regained a lot of traffic, but that’s coming to random posts that don’t attract my target audience. I’m basically getting more traffic, by focusing on bigger issues that more people are searching for; but that traffic for obvious reasons doesn’t need any of the services or offers I have. So my traffic has gone up, but my income hasn’t, yet.

That’s because at this point, a lot of my stuff is old, and I’m working on rebuilding all my sites, courses, and offers to be modern and useful. That said, almost everything I do will probably soon get replaced by AI. I’ve talked about AI cover design before, but there are already tools that can add text to images and do a pretty decent job. We might be a year out from “make me a thriller book cover with this title and this author name…”

So I’m scrambling to rebuild and relaunch things that will probably be useless soon – but I think that’s what every business is doing right now. We’ve had a lot of big technological advances in the past two centuries, but we’ve never ever been here before.

Writing books

I’m facing a similar situation with my books – I’ve had moderate success and I keep saying “when my series are finished I’ll finally make money!” but I took too long, and the income dried up, so I had to take a break. My immediate goal, is to rebuild my online platform just enough to take the next year off, and finally finish everything – I would have no confidence to actually do this, however, without the newer AI writing tools.

I know my stories, my characters, how I want them to go, and what will happen. But I also know myself: procrastination, lethargy, avoidance. AI tools might allow me to finish all my work, the way I want it, because it reduces the mental burden and investment of resources (time and financial).

People are quick to say that anyone who uses AI is a talentless hack or fraud (which isn’t new: I’m still upset about the War of Art’s infamous “To labor in the arts for any reason other than love is prostitution.”) I make no secret of the fact that I seek out or create tools that make it easier for myself – and other authors – to do our greatest work with less effort, stress and turmoil, and that I don’t believe suffering for our art is a necessary step to creative excellence.

But when people gloat that I, too, may soon be replaced… they aren’t wrong.

Can I create enough great content, in the next year, to be successful before I’m replaced by robots that can do it better? If so, what’s beyond that point? If not, what’s next for me?

I’m not as pessimistic or dejected as I sound. Despite the threat of AI, I find the available tools make it easier to complete the tasks I want or have to do. I’m making progress and doing things. That feels better than being buried under the weight of obligations and desires I have no motivation to complete.

As a quick example, theoretically I could translate my books with AI (deepL) and get them narrated (elevenlabs) – though there are few audiobook retailers that will accept them right now, I could make that free content available on YouTube or other sites. This kind of thing was previously very expensive, so few authors could expand their IP this way, but the bar has been dramatically lowered. But mainly, I’m just hoping to get through the 20 or so novels I need to finish all my series, and get them into the world.

I also plan to build out my fiction site, with art and articles about all the famous mythological gods and goddesses, to drive the right kind of traffic towards my novels. There are new opportunities here that I could not have managed to do before on my own (I’ve been teaching people how to market books for years but in most cases, it is not worth all the time and effort, because most books lose money).

AI assisted or AI generated?

I don’t think anyone can escape this question, so you might want to start thinking about it, too. Ideally, we would get AI to generate all the boring stuff we don’t care about. And we would get assistance with reducing our workload or managing the tricky, difficult bits so we could polish our best creative work.

But that line will be different for everyone, and get even fuzzier as the tools become more powerful.

Currently, most authors are against AI generated books or covers, but mostly OK with editing tools. The publishing platforms haven’t caught up with AI yet, but they are showing signs that they are getting ready and preparing for the inevitable flood of content, when anyone can easy write books from scratch with no skill of effort. Audiobook narration, cover design, editing and all the expensive author publishing services, might be something you can do on your own in the near future.

When you can get better quality services for lower prices, choosing to support humans (who can prove their humanity by being flawed and messy and imperfect) may be the right moral decision but the wrong business one.

I can’t speak for others, but for myself, I’m hoping to spend more time doing the things I want to be doing, and enjoying the creative process by making things I love. I want more of my best work, and given my human limitations that I can’t control, AI is proving to be a valuable, supportive tool.

AI apocalypse

I never get tired of sharing Roald Dahl’s 1953 The Great Automatic Grammatizator. Basically, all creative writers are offered the choice to sign away their style so that robots can keep writing books for them. It’s a devil’s bargain, but it’s the only way to keep food on the table. It ends with this:

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.

Dramatic, sure… but also fitting. I don’t think it’s controversial to suggest that AI is bad for writers and artists, and that they are probably feeling this same pressure (although, the false dichotomy between “selling out” by making commercial stuff and “keeping it real” by following your passion and ignoring the potential market, is a Romantic Ideal that continues to plague most creative entrepreneurs).

But the truth is… AI is only bad for established, skilled artists and authors, who have spent a lifetime improving and reached a point of success. For all unskilled creatives, who perhaps felt artistic but were told they could not write or could not draw and so were never encouraged of confident enough to pursue the arts to the point of proficiency, AI is kind of miraculous, like suddenly levelling up. It’s an unexpected boon, like the muse kissing you in a dream and giving you the World’s Best Idea.

AI will succeed, as long as it supports the many more than it threatens the few.

PS – I want to point out that Zeus originally sent Pandora to open the box as a punishment for humans playing with dangerous stolen tech; she was literally a robot (automaton) built by Hephaestus… but I have another article on that 🙂

Descartes’ Discourse on Method

* At this point I had dwelt on this issue to show that if there were such machines having the organs and outward shape of a monkey or any other irrational animal, we would have no means of knowing that they were not of exactly the same nature as these animals, whereas, if any such machines resembled us in body and imitated our actions insofar as this was practically possible, we should still have two very certain means of recognizing that they were not, for all that, real human beings.

The first is that they would never be able to use words or other signs by composing them as we do to declare our thoughts to others. For we can well conceive of a machine made in such a way that it emits words, and even utters them about bodily actions which bring about some corresponding change in its organs (if, for example, we touch it on a given spot, it will ask what we want of it; or if we touch it somewhere else, it will cry out that we are hurting it, and so on); but it is not conceivable that it should put these words in different orders to correspond to the meaning of things said in its presence, as even the most dull-witted of men can do.

And the second means is that, although such machines might do many things as well or even better than any of us, they would inevitably fail to do some others, by which we would discover that they did not act consciously, but only because their organs were disposed in a certain way. For, whereas reason is a universal instrument which can operate in all sorts of situations, their organs have to have a particular disposition for each particular action, from which it follows that it is practically impossible for there to be enough different organs in a machine to cause it to act in all of life’s occurrences in the same way that our reason causes us to act.