Sloptastic!

The malignant lie of "online community" and the cynical rollout of Sora 2

Community means a lot to me. I belong to communities that instill in me a sense of belonging and authentic connection to other people through shared interests, values, and goals. Many of my social and emotional needs are met through these communities, as they provide me opportunities to meaningly serve others. In real communities, I’m connected to people will I know will look after my family and me in time of crisis, just as I’ll show up to help them in their time of need. In so many ways, we define ourselves through the communities we belong to. Without community, we’re isolated, vulnerable, and depressed.

“Community,” like “content” and “ecosystem,” is becoming one of those words that the tech business seeks to redefine for its own toxic and cynical purposes. Technological developments are routinely touted as community-building. Sometimes they actually accomplish their lofty goals, such as the beginning of social media, before the profit incentive took over. As technology has quickly evolved exploded in wealth, we’re seeing more and more bullshit communities that don’t serve our needs, that divide and isolate us and even destroy real communities.

For example, crypto “communities” are often bullshit. The original idea of cryptocurrency involved an ethos of fostering a sort of community, albeit online and anonymous. It was intended to be trustless, decentralized, and democratic - a fair and sustainable alternative to a rigged and corrupt banking system. A community of token holders was supposed to ensure fairness by crowdsourcing policy decisions. It’s a beautiful concept. Instead, crypto has become a global scam inhabited by the most self-interested assholes on earth: a rigged casino where the house always wins, and a corruption device of infinite scale. The ideologues have fully left the building, but the scammers who remain continue to pontificate about the transformative power of community. Maybe it’s inevitable that money, or “stores of value” will lead to corruption. In the case of crypto, it happened almost immediately.

Then there’s the idea of the social media community. Tech oligarchs want us to think of socials as a community-building, as an extension of our real life communities we value. That’s right - we’re there scrolling our phones ten hours a day to be more connected to the world! In its infancy, social media did sort of create social connection and community in the real world.

All of us oldsters enjoyed the thrill of reconnecting with friends and past acquaintances on MySpace and early Facebook and making plans to get together in person. Then the algorithms took over, targeted ads and marketing, dividing us along political fault lines. Social media created the attention economy that keeps us constantly distracted. If we’re not fully present at dinner or in church because we’re sneaking peeks at our social feed in search of a quick dopamine hit, then we’re not doing community right.

Whether we’ll admit it or not, most of us are addicted to this sort of brain rot to some extent. I'll admit it. I scroll because I’m seeking something meaningful, seeking real connection through friends or music or whatever else. The dopamine payoff for me comes when I manage some semblance of real connection. What I get lately is ads for fitness supplements, dating advice, political rage bait, more and more AI slop. Why do I bother? I scroll looking for an off-ramp - a conversation with a person, or a rabbit hole to go down.

It feels like our efforts to market music and other forms of human creativity through social media are in fact efforts to break the spell and move people’s attention to something more substantial than passive engagement. There’s still some real stuff on those scrolls if people are willing to look for it, though real expression is clearly becoming deprioritized and endangered. Those of us looking for something real and inspiring will eventually fall out because the scroll becomes pointless and meaningless. We’re pulling the arm of the slot machine again and again, not for financial reward, but for actual human connection. For community. We’re trying to find something we’ve lost in an increasingly unlikely place. We’re all in the same online spaces, completely isolated from one another. Online community is becoming the literal opposite of true community.

We should never trust Big Tech when it says it intends to facilitate authentic human connection. It’s well established that Big Tech is an enemy of real community and authentic connection. Now that crypto and social media have failed to deliver on their promise of making the offline world a better place, along comes OpenAI with its highfalutin promise to improve our lives. It’s [technically] a nonprofit¹with a mission to “ensure that artificial general intelligence (AGI) - by which we mean highly autonomous systems that outperform humans at most economically valuable work - benefits humanity.” Just like Google’s old credo, “Don’t Be Evil,” they’re out to benefit - and by extension not harm - humanity. They’re just here to help!

What a relief, right? The odds-on favorite entity to rule the world through AGI has a stated mission to benefit humanity. I wonder how it’s going so far?

Last week OpenAI launched its latest video generating model Sora 2, along with a Tik Tok-style scrolling social app of 100% Sora 2-generated AI content aka slop. In the accompanying blog post setting forth the principles behind Sora 2, OpenAI stated that its aim is to “inspire people to create.” The post claims that Sora 2 prioritizes human connection, balancing freedom of expression with “robust guardrails” to prevent harmful generations. Among Sora 2’s offerings is “cameos,” whereby users can generate content with their own image or images of other users in what are undoubtedly fun and hilarious uses that don’t involve any sort of abuse or bullying, right?

So far it isn’t going all that well, that is if you care specifically about the potential for harm. The harms are visceral and immediate, and about to get a lot worse. It’s a disaster, a means to bully, misappropriate, and endlessly distract. Even though Sora 2 access is limited to select access code recipients, the slop created with Sora 2 is all over every social media platform. The videos look great! But the supposed guardrails are easy to circumvent. OpenAI isn’t operating on any sort of “first, do no harm” set of principles here.

Last week’s launch intentionally enabled its invite-only users to create with all manner of proprietary characters and famous personalities. The Reddit link I posted above is Jake Paul coming out as queer and offering a makeup tutorial. Not really funny at all, but people on social media have believed far less believable things. People shared the clip believing it was real. Online communities are actively blurring the line between reality and fantasy. We don’t know what’s real, and we haven’t yet developed the common sense to suspect that everything we see online is bullshit.

The roll-out offered an “opt out” opportunity for rights holders. Not a blanket opt-out, but rather an opportunity to object to specific uses. People had a lot of fun with it until the inevitable reversal by OpenAI. They’ve (predictably and likely as planned) reversed course on policy; however, people are still creating objectionable works in violation of IP rights.

OpenAI’s rollout strategy for Sora 2 is classic Big Tech playbook of launch first, legalize later. Get everyone addicted, then fine-tune the model for small nuisances such as laws and policies. Compare their rollout to rollout strategies of YouTube, Uber, and Facebook with its facial recognition and training on user photos. Each of these platforms knowingly launched with models that violated laws and legal principles. Rather than doing the right thing by acknowledging and delaying launch pending resolution of the legal issues, the strategy is to go straight to market and deal with the consequences later, if at all.

Sora 2 immediately went super viral, both because of the stunning quality of the outputs and the fascinating strangeness of making things we all know to be completely messed up and wrong. One YouTube video described Sora 2 as “the copyright Wild West.” OpenAI makes it so much fun to steal IP without any consequences! Feels a bit like…well…early Napster. Napster’s another copyright Wild West moment, one that ended badly for just about everyone affected. Particularly the random users who got sued by the major music companies.

Sora 2 is even more problematic than the other examples of the rush to market, because there’s absolutely no legal grey area in what they’re doing. There’s no safe harbor for a company that facilitates and enables the user infringement. It’s all infringement. Of course there’s a Fair Use debate playing out in courts over AI models using copyrighted works in their training sets; however, the fair use debate is with respect to the input, not the output. It’s well established that an AI output that infringes is just plain ol’ infringement. The rights for characters and public figures are far more complicated and multi-faceted than just copyright, including trademark and NIL rights. In short, allowing the sort of content they allowed at launch is unambiguously illegal on multiple levels.

Furthermore, getting us to focus on output infringement distracts us from the obvious fact that OpenAI is continuing to train its models on copyrighted and otherwise IP protected works. We know this because of the outputs. They expect us to be so excited about their product that we don’t mind that it’s all ripped off with zero transparency or compensation to the artists and rights holders. We need to keep our focus on the wrongdoing, and we need to hold strong to our values that are offended by this conduct by a major, potentially trillion dollar corporation.

When I asked ChatGPT about the Sora 2 rollout, it responded as follows: “Sora 2’s rollout is not a blunder - it’s the logical evolution of a Silicon Valley doctrine that treats legal grey areas as the cheapest marketing spend.” I agree to a point. I just don’t see the grey area here. I’m glad Open AI’s own product sees the problem, too.

This nonprofit supposedly looking out for the future of humanity in the face of existential AI threats is blatantly enabling infringement of intellectual property in pursuit of a viral moment to introduce a new doomscrolling brain rot app that will further isolate us with horrific AI slop in services of…creativity? Democratizing creativity? Let’s talk about creativity.

Sure, it’s creative to think of making a video with the cast of South Park read the script of The Godfather. It’s creative to make a video of your cat hijacking a jumbo jet. I suppose it’s technically creative to out some macho celebrity as a cross-dresser. It’s creative in that it involves some minimal amount of creative thought. Nevertheless, these scenarios require no real effort or skill, and they offer no insight into the inner life or values of the artist. It isn’t art; it’s amusement. It’s another ploy to dominate our attention with online curiosity and time-wasting scrolling. In short, it’s bullshit.

If OpenAI and others plan to build their AI slop/brain rot slop machines on protected works, they must do so with transparency, license, and compensation for every bit of content they use in the training. The rights holder and artist communities need to keep a hard line on this critical point. Outputs shouldn’t be “opt out,” or even “opt in.” They should be non-infringing, original creations - even if they’re not capable of copyright protection. In short, OpenAI, STOP RIPPING EVERYONE OFF. Stop making products that further divide and isolate us. Most of us are uneasy about the AI future; show some leadership. If Open AI needs money to cure cancer, fund your business in a way that doesn’t create new forms of cancer.

I don’t doubt that Sora 2 will create value in the world. It’s truly amazing that we’ve reached this point. I know a lot of marketers are excited to create quality ad and marketing content cheap n’ easy, and I’m sure that’s just the beginning of the value creation. Soon we’ll take for granted that anyone create anything they can imagine with a prompt…but it won’t necessarily make our lives any better and it doesn’t turn everyone into an artist.

Most of us value creativity and art because the connect us, build real community, and provide real insight into our ourselves and others. The promise of AI - the promise of OpenAI - has always been it would make our lives better, cure diseases, free us up to have more meaningful lives. You’re here to benefit humanity, not to throw rocket fuel on the raging fire consuming our attention, sense of community, self-reflection, and self-respect. Distraction isn’t joy, for most of us it makes our lives measurably worse and less enjoyable.

1. It’s complicated