It's been three months since my last post here on the blog. What can I say? I've been busy cooking up some delicious new graphic novel projects, and still can't quite talk publicly about them yet. Instead, you're getting a round of news and reflection from here at Scotty HQ.
THREE THIEVES: THE KING'S DRAGON Back In Print!
Business first:
As of February 3rd, the fourth THREE THIEVES book, 2014’s THE KING’S DRAGON, is back in print and in stores from the excellent folks at Papercutz.
As the series hits the halfway mark, it “dives unexpectedly into deeper waters,” according to Booklist, and “creates a riveting character study of a complicated, difficult, noble, and conflicted man, the likes of which one seldom sees in genre graphic novels for this age group.”
It boasts a new introduction by my friend Jenn Haines (owner of The Dragon in Guelph, Ontario, and former president of ComicsPRO) and among the extras is a continuation of the ongoing behind-the-scenes interview between myself and comics scholar Irene Velentzas that has wound through the new editions.
Still incredibly pleased that Three Thieves is back on the shelves, and in people's hands. Go get yourself a set, or replace the old library copies that have been falling apart.
AI DUMPSTER JUICE
The second item is also business, in its way:
I only barely use social media (I’m down to just Instagram, which so far there doesn’t seem to be a good substitute for, as much as I’d like one) and dislike wading into hot-button issues online. Everyone in the world is already doing that, no one ever changes their mind about anything (more on that later), and I’m generally pretty protective of my mental health. Besides, I’ve already got a job.
But about a month ago, I had to have a few conversations with arts organizations (AKA people who should know better) about posting AI-generated images online. In two of the cases, the images were submitted by outside users, and the organization hadn’t realized they were AI-generated (which is a bit of a concern in its own right). Hopefully this prompted those folks to adopt an AI policy (better late than never!) and, in one case, led to a good conversation about the ethics of AI.
At around the same time, there was apparently a meme going around that involved people prompting ChatGPT for caricatures of themselves (I seem to have curated my feed well, because I didn’t end up having to see a single one of these, I’m happy to say.) This kicked off of an umpteenth round of circular, online arguments about the subject of AI, with working-class artists across many disciplines saying “AI is bad for our industries, bad for the planet, and bad for our souls" and tech utopians/apologists insisting that the rest of us need to embrace a future in which they’re seemingly entitled to all of the world’s intellectual property in perpetuity. You know how it goes.
Make no mistake: I’m on the side of artists. Always have been, always will be. This is hardly the first time that technical innovation has led to exploitation. It seems like every few years, creative professionals have to defend their rights against some new thing that the law hasn’t yet caught up with. (Remember NFTs? Ah, the relative simplicity of 2022!) Still, many artists are, frankly, not very well versed in their own rights, or uncomfortable advocating for them, or might not be the veteran of as many of these battles as I am. As much as I may hate it, I’m a person who might be looked upon to provide some leadership on this issue.
Very few people seem particularly good at conveying why generative AI is unethical, in simple terms that can be understood by an audience of social media users who might have little idea how AI is trained, or how creative professionals actually make money. So, in light of all this discussion, I dashed off an all-text Instagram post (something I rarely do…it’s not really what the platform is for) that was intended to be simple, direct, and fit on one slide that could be easily shared. No highfalutin language about the nature of art and its intrinsic humanity, just my best ethical/legal, practical argument against AI, in an easily-digestible nutshell. It circled the globe a couple of times, and did so surprisingly quickly. Maybe you've seen it.
I avoided the environmental issues surrounding AI altogether. Sadly, that’s a failing that’s not unique to AI. A lot of things are bad for the environment, but I don’t shame people for, say, driving cars. I drive one myself (though not very often, and I try to offset that where I can.) I mostly tried to focus on equating intellectual property rights to physical property rights, which I’ve always found to be the best way to get “civilians” to understand IP issues. (Someone once told me they thought copyright should only last ten years. I told him to buy a house, and that in ten years my buddies and I would move in.)
I also avoided arguing whether AI is any good. People like to make fun of how AI can't draw hands, and how the images it produces are somehow weirdly yellow. That it hallucinates. That it’s racist. But those things have always seemed to me to be beside the point. Human creators also make bad art, that is sometimes unreliable or downright wrong. The important thing is that they actually made it. And own it.
Less effectively, I also decided to point out that some of the companies most zealously developing AI are owned/run by some of the worst people in the world. Even though everyone seemed to know exactly who I was talking about, and the connection between generative AI and the far-right has been made elsewhere (particularly well in “AI: The New Aesthetics of Fascism” by Gareth Watkins) I ending up regretting using the term “right-wing oligarchs,” because that gave the post a political cast that distracted from what was meant to be an ethical and legal argument, not a partisan one. I ended up deleting the original post for that reason. But — as proponents of AI data farming are always happy to tell us — once you put something on the internet, it’s out there. So it will live here, couched in a more considered and nuanced context, which I hope people will find.
Really, though, I think it speaks for how annoyed and exhausted the public is by the relentless marketing of AI in every aspect of our lives — not to mention the proliferation of political propaganda, revenge porn, chatbot-related suicides, etc. — that I got remarkably little pushback after posting it, at least in the corners of the web that I pay attention to. I got a lot of virtual high-fives and “hell yeah”s, as well as likes, shares, and reposts from people who were looking for a concise argument they could make to their boss, their Facebook-addled grandparents, their tech-fetishizing brother-in-law. That, of course, was the idea.
And at least one person told me that it changed their mind about generative AI. Which I was really happy about. Honestly, when was the last time someone actually changed their mind about something? Particularly on the internet. That guy is a unicorn.
For the record, I think the technology itself is pretty interesting. It brings us a step closer to a Star Trek-style holodeck future, which is cool. People who argue that artists are afraid of technology or change aren’t paying attention; usually the first thing that happens with any new invention is that creative people find a way to make art with it. Maybe that’ll eventually be the case with generative AI someday, too…though the results so far have been underwhelming, to say the least.
Some uses of AI are less offensive than others. I know some people use it to simply generate ideas, or to mock up work that they (or someone else) are going to make in a more legitimate way later. And I’m sure people will continue to figure out what it’s for (something the boosters never seem too sure about, for all their insistence that we all need to be using it.) We're often told that "it's a tool!", an argument they appropriated from digital artists. But so far, it's clear that it's mostly using us, not the other way around.
The thing working in favour of creative industries, at least for the time being, is copyright law. The U.S. Copyright Office has deemed that works generated with AI aren't subject to copyright, a decision that prompted Trump, beholden to the tech CEOs whose disinformation-filled platforms helped get him elected, to fire the head of the Copyright Office. I'm pleased to see, as I type this post, the breaking news that the U.S. Supreme Court, by declining to hear an appeal to a legal challenge, has upheld the Copyright Office's decision. It's doubtful that authors, publishers, movie studios, etc. are going to want to get too involved with works they can't exclusively licence.
And the law will continue to catch up. The settlement in the class action against Anthropic contained good news and bad news for artists, but will ultimately end up paying $1.2 billion to rights holders whose work was infringed in the training of generative AI, a welcome diversion of cash flow toward the people on whom the technology is actually built. There are many more such class actions still waiting to be litigated. Hopefully there will be more good news, and more legal precedent set.
With that legal framework in place, it seems clear to me — and to many other artists, whose ownership of their own work is at stake — that until rights holders have the ability to opt out of AI training data altogether, or to be compensated for the use of their work for that purpose, that there can really be no truly ethical use of generative AI. Even the most seemingly harmless application of it normalizes intellectual property theft on a global scale, and rewards and validates the thieves. It's certainly a terrible look for an arts organization of any kind, which was the original point.
In any case, you can be sure that any work of mine you encounter will continue, as always, to be made with my very human hands, brain, and heart. That, friends, is a guarantee.
