Am I Writing Like AI, or Is AI Writing Like Me? The Uncanny Creative Convergence

Do I write like AI, or does AI write like me? It’s a question I’ve started asking myself this year. There have always been “shortcuts” when it comes to writing and content creation, but the unveiling of ChatGPT in November of 2022 shifted the entire culture, calling into question the future of creatives across industries. The immediate response to the finally-realized “generative AI” dream went something like, “Great! We don’t have to pay for creative services anymore!” That sentiment was promptly disproved by the boundless faults and shortcomings of first-gen AI software, soothing the fears of writers and creatives everywhere.

I have no doubt that much of AI is here to stay—email sorting, case law research, brief drafting, contract review, and administrative automation for all, not just lawyers—but the consensus on AI-generated content has not yet crystallized. The AI itself is still in flux! How has emerging tech changed how we write? Is it possible to use generative AI in place of a writer or creative? Should businesses reasonably expect double the output with AI tools? I wrote professionally before ChatGPT, and I write professionally now. So here’s my State of the AI-Generated Union:

ChatGPT and its ilk are aids, not shortcuts for cranking out copy. They’re tools that work best and most efficiently in the hands of competent writers, folks with a firm grasp on language, the ability to edit, and critical thinking skills; people capable of doing most of the legwork themselves. ChatGPT can expedite content creation, but it cannot replicate human depth and nuance. Heavy dependence on generative AI for content and copy will absolutely hinder your brand. And if it’s not hindering you now, check back in a couple of years. What passes as acceptable now likely won’t in the future—the pendulum always swings back.

Sacrifices: Writing and Creativity Are Systemically Undervalued

Writing is a consistently undervalued skill (my own bias, obviously), and in professional settings, there has ALWAYS been top-down pressure to maximize output. It’s not an issue I plan to personally solve through this post, but it’s important to acknowledge that this is where the impetus to use ChatGPT for content arises: a pressure to churn out copy quickly and therefore at a lower cost. But creativity can’t speed up without a trade-off—quality for quantity. And that’s the reality we’re already seeing play out.

“Content” includes social posts like a BHM celebration on LinkedIn, and good social content requires seasoned copywriting. Content beyond blogs and landing pages encompasses adcopy, headlines, photo captions, descriptions, newsletters, swag slogans, event promotions, and more. It’s not that using AI to generate this content is a cardinal sin, it’s moreso that it reveals a fundamental misunderstanding of the creative process and immense legwork it takes to bring an idea to fruition—and a misunderstanding of what makes content resonate with people in the first place. 

Even short-form writing like social posts often requires more time and creative effort than a business wants to allot. When character or space constraints imbue every word with excess importance, you have to spend that time working through your ideas. Often there’s pressure to run with a first idea (never the best idea, as any creative worth their salt will agree) because Business with a capital B is all about spending as little time, money, and effort as possible, especially for work that doesn’t immediately result in ROI. The truth is that marketing is not sales. Good content is long-game. The creative process is like a fee your company must pay—skirt the fee, and the shortchanging will come back around. 

Sci-Fi, Fantasy, or Reality? AI’s Current Iteration

I can’t help but notice that we seem to have reached the age of the robot, and yet….here I am typing on my laptop at my job. Ex Machina-style artificial intelligence has yet to emerge, and the machines have unfortunately failed to take my job from me. The generative AI software that we’ve colloquially referred to as “artificial intelligence” is more similar to machine learning and algorithmic predictive text than it is a humanoid supercomputer straight out of a scifi. These platforms can’t be classified as “AGI” (artificial general intelligence), let alone ASI (artificial superintelligence), though they do “create new content.” The training that produces this new content relies on human-made materials—our books, our movies, our discussion forums, and our social interactions—and on human reinforcement and feedback. The goal of AI content development is to effectively emulate the characteristics that set our brains apart from machines. As AI evolves and potentially improves, we may become increasingly alienated from our intuitive ability to distinguish between human and machine. “Because AI seems to be human, it has the result of paradoxically making actual humans seem less so.”

For all my soap-boxing about how important human-made content is, I recognize that today’s internet is a deluge of the AI-generated. For all the tweets, headlines, bylines, descriptions, captions, posts, comments, and copy I passively glimpse on a daily basis, it’s likely half is machine-generated. There are all kinds of theories about the sprawl of AI content and the idea that it’s all sort of self-cannibalizing (dead internet theory is interesting)—and some research estimates that up to 57% of all web-based text is AI-generated or refined. 

Is AI Here to Stay? Considerations For the Second Half of the Chessboard  

The second half of the chessboard (Ray Kurzweil) refers to the point at which exponential technical growth becomes so significant that it leads to major societal shifts. I’m no technologist, nor am I trend forecaster, but I’d guess that the integration of consumer-ready AI software across our lives qualifies as a step into the second half—AI in your car, AI in your kitchen, AI on Spotitfy, Google, Outlook, Word; AI in the movies, AI on your billboard; AI in the house, AI on a mouse. It’s reached maximum Dr. Seuss levels. 

Many forms of AI are already so entrenched in our daily applications and processes that their survival seems sure. But AI in the arts—movie posters, graphics, scripts, books, and articles—is controversial. The top-down demands to use AI to cut down on the time, money, and effort have often resulted in degraded copy, low quality images, flat stories, and content that lacks depth and connection. You cannot automate humanity. Furthermore, heavy reliance on AI reveals more about user errors than it does display the limitations of the software. AI can act as an aid, one tool among many in a creative’s toolbox, but it’s only effective in the hands of a user who already possesses a strong grasp of language, critical thinking skills, and an intuitive sense of what resonates. 

Humans remain indispensable, not for pushing buttons and managing the AI output, but for creating nuanced, resonant content.

Innovation, Regulation, and The Environment

While this post doesn’t even touch the ethics of AI, it would be incomplete without mention of the severe, potentially extreme impact of this technology on the environment. Like Crypto, like blockchain, like streaming 4k on-demand, these models require immense computational power and add to the growing “need” to construct massive data processing supercenters. We know that new technology or emergent industries don’t disappear just because they’re polluting or destructive. They are “managed” and the harm is “regulated.” But, I do think we’re in the second half of the chessboard, both technologically and environmentally, and AI will become increasingly controversial as it continues to impact the world, the environment, our lives, our relationships, our language, and our humanity. From a business standpoint, you may not care about this larger, cultural AI discourse, but you can trust that it will impact your bottom line (and likely is already!).

The Cultural Pendulum

I mentioned at the beginning that what passes as acceptable now likely won’t in the future—the pendulum always swings back. I’m referring to the idea that any major cultural changes or societal shifts will always incur backlash. “Techlash” (coined by The Economist in recent years) refers specifically to the negative reaction to the “growing power and influence of large technology companies, particularly those based in Silicon Valley.” Silicon Valley is home to the OpenAI (ChatGPT) HQ. Techlash moreso concerns the profiteers of the tech world—in the case of AI, those making big bucks off our data or using our intellectual property to train software—than it does to dissatisfaction with the softwares’ existences or integrations or impacts. But, IRL, even before reading that word for the first time, I observed a more natural, social form of techlash that I believe will shape the future of AI-generated content.

Anytime we move forward—or sideways—there’s always a reaction. In an age of constant, exponential change, that reaction often looks like a longing for what’s more tangible, familiar, or human. These cultural, social, and environmental implications might seem unrelated to your plans and ambitions, but no business operates in a silo, and no man is an island. Companies may not be people, but they are run by people, and they exist to serve people. While the future is impossible to predict, staying grounded in your own humanity and the humanity of others is the best way to navigate it.

TLDR

AI can be a powerful tool for content creation, but it requires a solid foundation of writing skills, critical thinking, and the ability to connect with your audience. AI cannot completely replace human creativity and intuition. To truly leverage emerging tech, you need to master the craft it’s assisting you with first. Heavy reliance on AI has the potential to erode our natural communication abilities, our capacity to connect, and even our critical thinking skills. If you’ve made it this far, thank you for reading some of my un-optimized, 100% human thoughts—straight from my brain to yours.

Sharing is caring: