Fighting AI and remix culture is a losing battle

Many are worried about AI taking inspiration from their ideas. It couldn't be more misguided. The internet is a copy-paste machine - being copied is the goal...

Recent Tweet on the ongoing misguided mission of megacorps trying to neuter the internet (full story here). They still hate their own users remixing their works, giving it free marketing and awareness and instead would prefer people disdain them. Above ex of Disney, a ‘creative’ company run by bean counters and lawyers. Sad. No wonder Hollywood is so lost, it’s all the same story of beaurocrats running the world.

Obscurity is a far greater threat to authors and creative artists than piracy.

— Tim O'Reilly

I’ve already explained my rationale why you should embrace the use and reuse of your content and ideas — authorized or not — in many previous posts. There are benefits here on multiple fronts, especially since it’s something you’re not likely to contain. In a digital world it’s always more strategic to focus resources on that which you can control and ignore or try to leverage with automation that which you cannot. This is a difficult lesson the music and newspaper industries are still going through today, and they’ve mostly just succeeded in making their own users hate them (by believing they are pirates or criminals as the default, it couldn’t be more misguided).

We’re living through a digital renaissance, where copy-paste/remix culture is the default and to restrict or fight such activities goes against the very fabric of the web. 

Some thoughts on this, in no particular order…

You’re fighting a losing battle. If people want to steal your content or not, they’re going to. There’s no point in playing digital whack-a-mole with things like web scrapers. You can stop some of the bots and even use tools like Copyscape and scour the web for unauthorized use of your content, then pursue them to remove it. But even if you start to send takedown notices, there’s no guarantee that’s an effective tactic to stop this. You likely don’t have the resources to do this across every infraction, and even if you win it’s a Pyrrhic victory. People who know better don’t read scraper sites, and in time the true sources of content become apparent. The web’s users (the ones who are worth anything, at least) are smarter than you think. And your memes? Those are going to get re-used too, whether you like it, or not. The tradeoff of the internet necessitates this is the case, but it’s more than worth it. Value and reputation still assuredly accrues to you.

Search engines know what the original source is. Google is smart, I’ve never seen them rank a site that scraped my personal content higher than the original post and even if that happens it’s fixed very shortly. If this is happening to you, you likely have technical SEO issues or some strikes against you for some reason and you should work to resolve it. Sites that are made up of purely duplicate content ultimately always fail from a search standpoint, and the engines are only getting smarter at sorting the wheat from the chaff. Much of the web wouldn’t even exist today if this was an existential concern. If you’re producing anything at all meaningful none of this is something to worry about. A pure clone of the WSJ wouldn’t get much traffic, for example, and definitely wouldn’t get any paying subscribers. Brand matters.

You’re never going to be perfectly credited on social, and you shouldn’t care. People ‘steal’ memes all the time and use copypastas etc. So what? Watermark your images if you want credit (which web comic creators do, and probably 99.99% of people never remove), but why even put things online you don’t want shared? If what you’re doing is stylistically interesting over time you’ll capture enough attention and enough of a reputation to have leverage to sell. If you’re already popular enough to get copied, you already won here and should have better things to do. Imagine if Mr Beast spent his time worrying about the countless Beast clones instead of just working on his next idea. You just can’t do it. He understands the internet, and so wins in an outsized fashion.

Why not put scrapers to work? Add something like an RSS footer plugin to WordPress or code something similar into your site, this way when your content is scraped, a link to your original post is scraped along with it. Now you just turned it into a positive: if others stumble upon your site away from the main source, it’s free promotion for you as it will contain links back to the canonical version. Even if you’re not using this plugin and others are scraping your content and keeping the links within posts, you win. None of these sites get any traffic of consequence anyway, it’s just not something to lose sleep over.

Digital success is all about making copy-paste easy. As an artist I’ve been having great results with “letting go” of my work through tactics like purposely seeding it on file sharing sites free. Only by making copy-paste easy was I able to broaden the reach of my art. Copy and paste is the best social media tool, no wonder big music run by lawyers and bean counters without the faintest understanding of culture or the internet wish you couldn’t do that. They’ve been wrong on all this for 20 years. Anyway, for those reading, why not make this as easy and encouraged as possible in a way where you’re credited how you’d like? And in cases you’re not, you’re actually still winning here.

AI can’t steal your brand, just market your style. And this is a great thing! Your style belongs to you, and even if people were asking AI to create something ‘in the style of <blank>’ it just means you already won. This is basically the Pixar example at the top of this post. Homage is a sign of respect and this is an incredible free marketing tool you were just gifted, free. If you care about stopping this from the early days, you’re never going to make it very big. If you already succeeded and want to stop people from having fun with your creativity and being a part of modern remix culture, you’re so incredibly lost as an artist I think you need professional help.

Memes and gifs are how content gets discovered. This is really one of the only areas big media wasn’t able to clamp down on due to fair use, but I’m betting you initially they would have if they could. Yet I’ve discovered, then purchased, many movies, TV shows and albums from internet users having fun with the IP of large media companies. Brands with marketers who aren’t asleep would kill to be able to create such things organically. While meme culture thrives in the grey zones of intellectual property, it’s extraordinary to me the benefits from this aren’t appreciated more widely and modern media lawyers don’t ease up on how they view similar things. Again, the real risk is not being included in the creative parts of the internet. In an infinite content world, the risk is always obscurity.

Likely, those ‘stealing’ your content are not inherently evil. Most probably just don’t know better. And majority of them are more than happy to include a link back to your original post or credit/citation if you ask. It is my belief that most want to do the right thing and work with content producers. That’s why I try to actively encourage reuse through a creative commons license, which encourages others to use (and link) back to the original source. Of course not everyone follows the rules, because not everyone is educated on citations, but most do. In reality almost no one is an outright pirate unless the user experience of your media is so dismal they’re forced to.

As Techdirt pointed out in 2005, none of this is a real problem. And their blog has millions of subscribers, so it seems like all those scraper sites and copy-pasters didn’t stop their success. They focused efforts on producing important stuff as opposed to worrying too much about thieves. Here’s a quote from them from years ago which is still true and reinforces the last point:

However, when that’s happened to Techdirt we’ve discovered two things — and both suggest that all this debate is a waste of time. First, when we shoot off a quick email to the sites asking for proper credit, we almost always hear back with an apology, from some “new” blogger who isn’t quite sure how it all works, and they usually fix things right away and start giving us credit. It helps, by the way, that we first thank them for finding our content valuable enough to reuse, and then ask nicely for proper credit. However, much more importantly, the sites that (1) don’t credit properly and (2) don’t respond to such emails almost always disappear within a month. Why? Because no one reads them. Who goes out and finds some nameless site that’s obviously reposting content?

You’re putting your work into a medium where if it’s interesting, it’s going to get copied or remixed. Digital content producers and media outlets need to make peace with this.  There is no efficient way to stop it and it’s likely not worth the effort. Isn’t the whole point of publishing digital content to have your ideas shared anyway? Wikipedia’s content gets copied all the time. Yet they’re still one of the most popular websites on Earth, by a pretty wide margin (4-5 billion visits per month). What does this mean? Keep your head down, stay focused and you’ll become referential, users will credit you, share your ideas, and then you win.

It’s not really about your content as much as it is about the context. That’s why I think people are more interesting than brands of media here, especially as many larger institutions become empty shells of themselves. I’m less interested in where that content is published and more interested in who created it. And it’s obvious that true scraper sites or banal meme accounts don’t have real people behind them or care to be a part of communities or industries, they’re just extractors. The terminal value of all these things is zero, and you should have better things to do with your time so just ignore them.

The Streisand effect is real. If you try to get content pulled, you’re merely just going to draw more attention to it. Likely different motivations here, but still the same effect. You’re just increasing the value of something and drawing attention to it by attempting to censor or remove it.

Wrapping up…

I strongly believe the main aspects of AI image use will be memes and replacement of stock photos. These things lend themselves to it. So, you want to be included here and want them trained on your data, it’s basically just free marketing for brands who don’t constrain themselves.

Fun AI meme going around, I reiterate memes and fun are going to be the main use of AI for awhile, unsure why you’d turn your back on it.

I know it’s perhaps frustrating to some of you, but the reality is we live in a copy-paste world and AI tools are just the latest iteration of this. It doesn’t matter what type of creator you are, if you can find a way to embrace copy-paste, you put yourself in a far more advantageous position than those who fight it. DRM and other forms of control are failure, basically a cancer on the internet. You can waste resources taking legal action or sending takedown notices to your own fans and make them hate you, or strategize ways to put them to work for you or your brand and leverage them.

The choice is up to you – however I’ll leave you with this piece of advice. There are limited amounts of hours to spend in online marketing, and it pays to make them count. Is worrying about scrapers and playing an infinite, unwinnable game of digital whack-a-mole really the most prudent use of time? The answer is pretty obvious to me. But I’m just a marketer and creative person, not a lawyer. So I want to see you win and make more interesting works, not waste your life litigating or anxious about something that doesn’t really matter.