We need to talk about AI

We need to talk about AI. I know, I know. BELIEVE ME. I know. But as a creative who frequently exchanges their art for money I think it's important to state where I currently stand, and I have more thoughts than a Bluesky thread can reasonably hold.

A quick note on what I mean by "AI" - I'm using a contraction that's become commonplace but actually can mean at least a couple of different things. A lot of what's being called AI in marketing copy is just old fashioned machine learning. This covers things that have been around for DECADES like predictive text or photo editing tools which shortcut manual work. Things that even the staunchest AI critics have used without compunction and will likely continue to use. What we need to talk about is specifically Generative AI - the stuff that immitates human-made art.

The first thing I'll say is that I've never used gen-AI in my games and I never plan to. There are a couple of reasons for this. One is that I don't find it interesting or joyful to do so. I make art because I need to for my own wellbeing, not because I want to be seen as an artist (although being completely honest I do also want that.) I think gen-AI "prompt engineer" artists are confusing the two. If you're not interested in the process of making art then you're just LARPing as an artist for the fantasy of it. This feeds into the second major reason I don't use gen-AI in the games that I sell: I don't see why on earth you would pay me for that. Some gen-AI optimists want to say that AI is "democratising" art, making it accessible to everyone. Even were that true, which I would debate, no one needs to pay the "artist" who generated it. They could just generate nearly the exact same thing themselves.

When we pay for any object the value it provides can be seen as a function of two variables: access and convenience. I can't make my own laptop, so I buy one - that's access. I can make soup, but if I'm out of the house and hungry it's much easier to buy some from the nearest soup vendor - that's convenience. I could learn how to make the delicious chow mien from my favourite Chinese takeaway, but it'd take time and effort that I don't want to invest right now. I just want my noodles. So it's access and convenience that makes me place an order. If I have convenient access to something I don't want to spend money on it. The jeans already in my wardrobe, the tea on my kitchen counter, the words I'm currently writing, the thousands of AI images I could create in an afternoon if I so chose.

Showing up in a marketplace full of artists who've toiled to build skills and produce art, then scaled a wall of their own insecurities to put a price tag on it and put it in front of consumers with something you "made" by whispering your ideas to a computer is WILD. And if you don't tell the consumer that what they're paying for is already conveniently accessible to them... that's what I'd call a scam.

I'm aware that in reality a lot of this stuff isn't 100% gen-AI. In the TTRPG space it's most often words by a human with art by a machine. Art is important for selling games, so one argument says that a writer without money for art or access to artistically talented collaborators would otherwise not get to sell their work. But free and/or cheap stock art exists. Also, I still sell three of my early games which were illustrated by me or my husband. We are not professional artists. The results are not polished. But I think they are charming and consumers don't seem to mind at all. Our non-professional illustrations are enough to convey the vibe of the games and those games were enough to get me a little pot of money to pay for my first convention stall, which carried me to my first crowd-funded project, which allowed me to pay a professional. Don't be lazy, you'll be fine, I guess is what I'm saying to anyone in this position.

How do we stop the gen-AI scammers?

I am online. I curate a timeline of good people who are passionate about human art, so I see you out there demanding that all gen-AI stop immediately and that all TTRPG marketplaces commit to an outright ban. I sympathise and know in my bones that that's not going to happen.

Why? Because I've been vegan for over 10 years now. I know how much energy, time, and conviction it would take to get even my local independent cafes to stop selling bacon sandwiches. It's so much, you guys. Oh god, so much life-consuming effort. And just typing that I needed a moment because I recently saw a video of a pig who refused to stop cuddling its human friend. You, dear reader, most likely eat pigs and I think that's fucked. I don't often say so though, because I know saying so won't stop you and then it'll be an awkward THING in our friendship. You won't engage with the point I'm making, you'll just remember I made you feel bad and like me a little less for it. Now I'm a buzzkill and you're craving bacon to cheer you up. This, I have learned, is how humans work. Even really good ones like you. I do the same with other things myself. Neither of us is a perfect moral reasoning machine and we don't all have the same values to start with.

How does this relate to gen-AI? Well, like meat, gen-AI is easily accessible and convenient and people seem to enjoy it. Some estimates place the environmental impact of one hamburger higher than an entire year of ChatGPT use ^ref. The alternatives take more work. Sometimes a little, sometimes a lot. And there will ALWAYS be a group of people who just do the convenient, accessible thing every time no matter what you have to say about ethics. And if they're buying, someone is selling. No matter how much suffering, human or otherwise, the conveniently accessibly option causes. So far so capitalism. But wait!

Find artists, pay artists

I'm aware that I sound jaded. The thing is, I'm actually an optimist. I don't think gen-AI will replace human art. I think it could, maybe, force us to reckon with the proper place of art in our culture.

The labour of artists has been devalued steadily over our lifetimes. Gen-AI is only the latest threat in a long line of obstacles including arts funding cuts and cultural indifference. The general non-art-making public have been convinced for a while now that entertainment is cheap or free. That you should get all the music in the world for the price of a Spotify subscription. Ditto television and film with streaming services. Digital art is only a right-click-save away from being yours and now AI will even help you remove any pesky watermarks. All the while we still crave art. We need it. We surround ourselves with it. Look around you right now and consider how many things your eyes land on that required someone's creativity. Is it nearly all of them? Now imagine finding out that an AI did all that. How does that feel?

Here's where the gen-AI-as-meat comparison diverges: people do actually enjoy meat. Many like it better than the alternatives, and so the only reason to overcome the convenient access problem with those alternatives is moral conviction. Gen-AI is not as satisfying as human-made art. It's not moral conviction alone that will drive people to seek out the good stuff. People want to see and hear and read things made by people, and I believe that we'll only become more conscious of this as we're increasingly surrounded by the alternative. Because, again, we value things more when they aren't conveniently accessible. In a world where "I didn't use gen-AI for this" is the exception rather than the rule, people will naturally have more admiration for the exception.

Up until a couple of years ago if there was art then of course an artist made it. We didn't need to think about them and their process and their time. Now we will have to ask ourselves those questions if ever we want to find artistic work that communicates something new about the human condition. Photography didn't kill oil paintings, it just gave us more questions to ask when we see an image that could be either. And which answer would make us say "wooow" and lean in? What makes us excited? The painting, because of the skill involved. I'm not saying photography is just like gen-AI, to be clear. Although we experimented with the new tool of the camera, initially only used to replicate, until we found ways to make interesting work with it. Likely we'll do the same with gen-AI once the initial excitement dies down and we understand its limitations better. I don't personally believe we're close to creating sentient life with gen-AI, but I'm not about to make this essay longer by going into all that.

Finally and to be clear I'm absolutely not saying artist are unharmed by gen-AI. For my day job I'm a software developer and the big fear in our industry is the loss of entry-level jobs. Right now Claude needs supervision by an experienced developer to get reliable results, but the same is true of the majority of junior devs. Even with 20 years of experience I myself may be obsolete within a few years if Claude keeps improving. I see a similar thing happening in the arts with things that previously provided the only sort-of-kind-of stable bread and butter for artists: stock photography, illustration, copy writing etcetera. It's important that those of us who are ahead of the curve in caring about this keep seeking out human artists and paying them money. We must be conscious when we consume art of any kind for free - why are we okay with this? If we're not, can we make reparations somehow? We must continue to push for a world where artists don't NEED to make their money doing the less interesting jobs that gen-AI can do. Where they have grace and space and time to make art for art's sake and to deeply learn their craft without being financially punished when they make work that isn't wildly popular. Because we know popularity doesn't directly corrrelate to artistic merit, and because real art isn't fast or free, and the misconception that it is didn't start with gen-AI.

Fighting the tide of gen-AI is a valid moral choice, but I worry it's more likely to burn us out than stop the tide. Instead we might use our energy better by pushing for ways to differentiate between works with gen-AI and works without and loudly, proudly championing the human art and artists we love.