The Worst Use Cases for Generative AI That Are Already Mainstream

The past 5 years have been weird. We had a pandemic that sent everyone inside for a couple of years. In that time, people got a lot more comfortable with technology. Then, the earliest generative AI models were released, and we've uncritically accepted them as the next big advancement. Sure, people are banning their use for trivial reasons like cheating and plagiarism, but no one is really pushing back on the tech in a meaningful way. For example, how do these tools harm labor? What about privacy? What are the goals of the companies making these models? Are their motivations ethical? Do they care about who is affected? Why is this tech being made? Has anyone considered the externalities? Now, I can't say I have the answers to these questions, but it bothers me that no one is asking. Like, did we even consider how people would be using them before we released them to the public? I don't think we did. After all, this article looks at several use cases of generative AI that I find dystopian at best, and I want to talk about them. Manmade Horrors Beyond Your Comprehension As a generative AI hater, I've spent a lot of time talking about the tech and its issues. For instance, I think the idea of having a machine which allows us to outsource our critical thinking is an existential threat to humanity. Yet, tech bros see generative AI (and the belief that it will lead to AGI) as the next step in human evolution. Perhaps you should be very wary of anyone with that mindset, such as Sam Altman who speaks of AI like a deity. That said, I am somewhat tired of reading about these technofascists and thought it would be more fun today to have a bit of a laugh at how folks are using generative AI. Though, I suppose it's more depressing than funny, but I enjoy laughing at the absurd. Education I think it goes without saying, but probably the number one usage of generative AI, especially text-based tools like ChatGPT, is education. Of course, I don't have any stats for this, but I experience it every single day. Students use generative AI tools to do all of their thinking for them. Given the extent of disruption caused by tools like ChatGPT in education, educators have seemingly given up on fighting it already. At my university, there is already a new class on how to use generative AI tools. In addition, the university already has a new initiative to embed "AI" in every course, which is perhaps unsurprising given that this is the same university that had a crypto bro as a graduation speaker. To me, forcing generative AI into the classroom is hysterical. After all, anyone who has been to college knows that professors are not trained to teach. They have almost no understanding of pedagogy, and the vast majority of them just read off of slides. Therefore, rather than spending resources on training professors to be proper educators, we're just going to turn them into ChatGPT merchants. Certainly, if I had my say, we would be overhauling the way education works anyway. We would prioritize more tangible outcomes, like learning, over grades. We would also prioritize the social and communal aspects of education through active learning. With the way technofascism going, however, I wouldn't be surprised if we just moved to a learning model where every kid sits in a pod and is fed individualized generative slop. Therapy One of the more disturbing use cases for generative AI I've seen floating around is therapy. It's kind of funny because I often see folks use "woke" language to justify its use in this context. For instance, people often cite the cost of therapy as a barrier to seeking out a therapist, and I get it. But, it's like doing a DIY fix on the sidewall of your tire instead of replacing the damn thing. You're doing a lot more harm than good. After all, there are already a handful of horrible suicide stories that are attributed to chat bots, one of a child and another of an adult. Not to mention that tech like this exacerbates the loneliness epidemic. Why wouldn't it? It simulates endless social interaction without pushback. To make matters worse, cost isn't the only reason folks are seeking out chat bot therapy. If you feel like a burden to those around you, why wouldn't you turn to a chat bot to offload your problems? The value proposition is built in. Just think about the idea of having a simulated social interaction without all of the mess of reality. ChatGPT is never going to push back in any meaningful way. Even if you coach it to, it can't hold you accountable in the real world. It's fundamentally spewing slop with no repercussions. Childcare As a newer parent myself, the idea that people use generative AI tools to entertain their kids really disturbs me. Now, the conversation of parenting is always a dicey one. We've all sort of agreed that everyone gets to do it however they like without judgment, but I'm going to draw a line for myself here: you should not be using generative AI to entertain your kids. The story that really set me off was a post from Reddit where the author got tired of talking to their kid about Thomas the Tank Engine. In that story, the parent's solution was to hand their kid off to ChatGPT, which then was able to "talk" to the kid about his favorite show. Now, there are a couple things I need to state up front before I comment on this story. First, I do not trust anything on Reddit (or the internet) anymore. This story could easily be fabricated or even written by ChatGPT itself, but I don't think that takes away from my commentary. Second, I am not just a generative AI hater. I think most current technology is damaging to kids, and it's why my kids will never be iPad kids (and I certainly won't be falling for the "but kids will be behind if they aren't exposed to tech" narrative as a literal software developer and educator) . With that said, this story is so disturbing to me simply because it's giving the kid a false sense of how social interaction works. Tools like ChatGPT are not people and do not behave like people in the real world. If a kid somehow becomes convinced that they can treat a regular person as a choose-your-own-adventure game, they're going to have a hard time adjusting. While the social implications are my primary concern with using generative AI for childcare, there is a laundry list of dystopian implications. For example, you can't filter what the chat bot is going to say to your kid, right? Say hello to the new alt-right pipeline. After all, we already have grok denying the holocaust. In general, reading through that thread is a real treat. While the majority of folks were obviously grossed out by the idea of ChatGPT childcare, there was still a vocal minority trying to moralize it. For instance, I saw one comment to the effect of, "it's better than talking to pedophiles on the internet like I did," which is a really strange argument. Of course giving a child unfettered access to the internet is bad. That's like telling someone whose kid smokes, "well, at least they aren't doing crack." Sometimes I feel sort of silly even pushing back on ideas like "your kids probably shouldn't have unfettered access to ChatGPT," but then I think about all the harm social media has done to teenagers. Just like social media, you have to be at least 13 to use ChatGPT, and you're required to get parental permission until you're 18 (of course, I have no clue how that would be enforced). Ultimately, maybe handing your 4-year-old off to ChatGPT is a bad idea. Though, you should absolutely educate your kids about it. An outright ban just means they'll be exposed to it outside the home under conditions you cannot control. Also, it's totally okay and even good for your kids to be bored. You don't have to slap a screen in their face or make use of every modern convenience to solve the "problem" of boredom. Creation I'll round out this list with one of the worst use cases for generative AI: creation. From an educational perspective, creation is often the highest form of achievement. After all, it's the peak of Bloom's taxonomy. To get a student to the point where they can create something is extremely fulfilling. As a result, when tech bros are touting that they've built a technology which can "create," it's a bit of a slap in the face. After all, creation is a uniquely human thing, as I've ranted previously. Yet, people like Sam Altman think it's possible to make a machine that can outperform humans. To me, this is a very frustrating idea. Could a machine be built that is more intelligent than humans? Sure! Can a machine ever experience life like a human? No! I would welcome a machine that could solve some of our problems through its infinite wisdom. I would gladly take a machine that could cure cancer or solve climate change. Would we ever act on that knowledge? Probably not. We don't act on the knowledge of the current best and brightest scientists. How does such a machine even help us? Even then, that doesn't seem to be the goal of such a machine. These weirdos want to so badly merge with machines that they'll destroy whatever currently exists to get there. Side note: I was thinking about this idea of the singularity a bit more deeply, and it really is eugenicist rhetoric (no, really). Even some of the folks in some of the AI threads I was browsing described having "normal" children in the future as being akin to child abuse. Really? But, giving ChatGPT to a child isn't? I can understand ignoring modern science, such as not getting vaccinated, as a form of child abuse. But, surely humans can more or less exist as they do now (and have) without the need for augmentation. Animals seem to be doing just fine. Also, not to be that guy, but a lot of these weirdos seem to orbit Destiny's subreddit. I don't know why that always surprises me. He cultivates quite the community. Perhaps debate bros are just losers like we all thought. Why else would they be trying so hard to augment themselves? Don't believe me? Look at the internet right now. All the "articles?" Yeah, bots. All the "comments?" You guessed it, more bots. All the "art?" Yup, more bots! I use quotes here deliberately when I'm talking about tools like ChatGPT. None of these generative AI tools are human, so we need to stop personifying them. What an LLM spits out is not creation; it's computation (and yes, I know the previous speech pattern reads like AI—a fucking human wrote that out first. AI doesn't get to take credit for it.). And it absolutely drives me nuts when the opposition positions this framing as "human exceptionalism." The argument itself is sort of silly. Plenty of people are smarter than me, better than me at sports, better artists than me, etc. I am not bothered by this. In fact, I am not even bothered that tech is better than me. I'm not mad that a calculator is better at computation than me. It was never about performance. That's an extremely narrow framing that I think probably comes from this western idea of meritocracy. Finally, losers can pretend to be good at something because they can prompt an AI to do it. I am not impressed by this, nor do I care. What bothers me is that a machine can never experience life. It knows no pain. It knows no joy. It knows no loss. Therefore, the best it can do is imitate the products of those feelings by harvesting the entire human experience. And, for what? So, it can produce some shitty pictures and articles? Is that really all we could come up with for a use case? Internet Points Okay, I'll share at least one example that is kind of funny. Has anyone else noticed how many people just defer to chat bots now in conversation? I saw this one thread where a text was incomprehensible, and one person wrote: I used ChatGPT to clean up the message Surely, anyone could do this, right? Why copy the output of a chat bot for anything other than internet points? Are we supposed to be proud of you for that? Not to mention that using ChatGPT in this context seems distasteful. If someone can't understand what their partner is trying to communicate, why aren't they talking to them? Why are they turning to Reddit? And, rather than giving reasonable advice like, "I don't know. Have you tried talking to them?," folks are using a third party software to repackage what a real person said without any of the real world context. It's all so dystopian. I really hope people aren't using tools like this to decipher texts from friends and family. It's bad enough that therapists are stuck helping clients decipher texts. We really have lost all social skills. Race Science Recently, I saw a thread of people sharing their DNA breakdown, their age, their body type to ChatGPT. Then, they asked it to guess what they looked like. Now, I don't know how to tell anyone this, but this might as well be modern race science. I would hope at this point that people know that ethnicity doesn't directly map to appearance, and that people of the same ethnicity can look wildly different. In fact, even if it were true that appearance could be exactly mapped to ethnicity, the percentages you get from companies like Ancestry and 23andMe are just estimates. Likewise, I can already see weirdos co-opting this idea to promote more eugenicist rhetoric. After all, the models themselves already feed on racist material. Surely, when someone with African heritage does this it won't result in a racist caricature, right? Not only that, but it seems odd to just dump all your deeply personal information into a chat bot, right? Why not give it your social security number, blood type, and address? Relationships Just watch this. If you come away with any feeling other than an existential crisis, you're already cooked: https://www.youtube.com/watch?v=7OTv6ZSP18c&pp=ygUMYWkgdXNlcnMgY2Jz0gcJCb4JAYcqIYzv Reactionary Resistance? As a kid, I grew up with Gen X parents. Often, when I hear stories about the things they and their peers did back in the 80s, I feel like a prude. After all, they're sometimes called the "free-range" generation as their parents left them unsupervised most of the time. Given their life experience, a lot of them went on to be tough parents in an effort to provide their kids with more structure and guidance. Interestingly, despite being Gen X, my parents actually gave me a ton of freedom. In fact, they treated me like an adult from a very young age, something that I'd like to carry through in my own parenting. Unfortunately, the freedom I grew up with was very different from the freedom they had. Rather than being free to roam the streets, I was free to roam the internet. As a result, I grew up with almost completely unfettered access to tech, though most of my time was spent playing video games. To put things in perspective, I had Xbox Live at ten years old. By the time I was in high school, I had spent quite a lot of my time playing video games, whether they were online games like Halo 3 and World of Warcraft or single player games like Pokémon Ruby and Super Mario 64. Despite having that much access to tech, my parents seemed to have a good gauge on what would actually be detrimental to me (or maybe we couldn't afford more). As a result, I wasn't given a cellphone until I was 15, and I only received that cell phone to contact my parents when I needed to be picked up from after school activities. Shortly after that, I created both my Facebook and Twitter accounts to stay more connected with my peers. Looking back, of everything I had access to, social media was by far the worst thing I could have consumed. It is something I have been battling with since I was literally 15, and I only recently got rid of both my Facebook and Twitter accounts at 31. Right behind social media has to be my phone. It's something I used to constantly use to check social media, but now it's something I hold onto just to pass the time. Often, it just gets in the way of having a genuine interaction with someone. I can't tell you how many times my wife and I repeat ourselves to each other because our phones are just a constant distraction. Funnily enough, I find the phone addiction is even worse with Gen X family members. They legitimately never put their phones down and often just throw on a movie or TV show for background noise as they scroll away. Or put another way, sometimes family get togethers are a little painful on social front. Looking forward, I can only imagine these issues getting worse with the advent of generative AI. It's already bad with family members sending along AI generated memes. I know it's only a matter of time before my peers start doing it. Hell, I think the first time someone I know tells me that they asked ChatGPT for something will actually kill me. So, all of this is to say: forgive me for not trusting a tech company when they say that generative AI tools like ChatGPT are the future and are good for us. The last 20 or so years have brought us "convenience" at the cost of the fabric of our society, and we're just happy enough with the slop to not complain. We have the power to resist, and we should leverage it. On my way out, let me just really quickly illustrate why I think generative AI is so silly, even as a tool. Right now, I can add an image to this post. Typically, I pull my images from royalty free databases like Pixabay. After all, the author consents to my usage of that image there. Now, when I add a featured image, these days there are tools that desperately want me to use them to generate an ALT tag. Supposedly, they will look at my image and generate a description, but I've never used this service and never will. Instead, I write my own ALT tags. For instance, on this page, I have included an image of a stream. The small joke being that the stream is a play on the word, "mainstream." I include that in my description to give visually impaired folks a little laugh and as an easter egg for folks browsing the site. Can generative AI do that by just looking at the picture? Can it infer intent? Of course not. If we are moving in the direction of using generative AI for productivity or convenience, we're going to lose these little things. Everything is going to be more generic, more corporate. Do we really want to live in that world?