Will Artificial Intelligence End Writing As We Know It?
The robots are coming for our language, and we're not ready.
Disclaimer: I am not an AI expert. I do, however, closely follow technological trends and trajectories. I also love words and language, and I’m a forever-student of what it means to be human.
Unless you’ve been living under a rock, you’ve probably heard about ChatGPT. You may have even experimented with it yourself. It’s a remarkable technology, capable of passing the bar exam within the 90th percentile, as well as the LSAT, GRE, and other standardized tests. It can problem solve, process up to 25,000 words at a time, and write college-level essays by being provided with a simple prompt. It’s leaps ahead of anything we’ve seen before, and its capabilities are growing by the day.
There are an almost infinite number of practical applications for ChatGPT. Ask it to write an email excusing you from work because of sickness, and it will craft a more convincing argument than you could likely come up with on your own.
If it can do that, for free, why would you waste time writing something similar for yourself? In fact, wouldn’t it be a waste of time to write almost any business-related e-mail? Take this one for instance:
Not bad for a first draft. But of course, ChatGPT is not just limited to business writing. Tell it to be witty, and it will try its best to do so. It’s sampling from terabytes of information after all—including many of the wittiest writers out there.
If you’re struggling to find a date, why not have ChatGPT compose your dating bio for you? Why not have it send condolences to your friend who just lost a parent? We all wish we could communicate more efficiently with less effort.
Well, now we can.
That’s fun, right? So what’s the catch?
Yes, OpenAI, the company who created ChatGPT, will use your prompts to further improve their product. But every tech company uses your information like that. Will they sell your data to advertisers? Probably, but again, so does Facebook, TikTok, etc. In my view, that’s not the main problem.
We are quickly approaching a point where it will no longer be possible to tell the difference between writing done by a human and writing done by a machine. In many ways, we’re already there. How, then, do we interpret the information we are reading? Can we still read a heartfelt note from a friend with the warmth and sympathy it appears to contain, or must we assume it was compiled by a bot?
If the person sending you the note intended it to be warm and heartfelt, does it even matter that they did not pen each word themselves?
Be honest. How invested are you in a typical work email? If you are like most, probably not very. Now consider how much less interested you would be if they weren’t written by your coworkers, but by a machine.
I’m yawning already just thinking about it.
The impending robot takeover of the written word will inevitably make our eyes glaze over our day-to-day communications. Why care about what a message says when you know it wasn’t written by a human? Our shorter attention spans are already prone to ignoring writing in favor of listening to audio or watching video—and we still can’t keep up with the impossible flow of information that comes our way.
Will we stop believing that what we read is real, even when it comes from trusted sources?
What will it do to our species when we no longer need creativity, emotion, and empathy to communicate?
It’s true that humanity has (mostly) been able to keep up with technological advances, and it’s possible we will be able to weather this one too. I don’t think it’s a bad thing that we can reduce the amount of effort we put into mostly meaningless tasks. Getting rid of busywork is a net-win. It’s when we cease to use our imaginations or no longer find meaning in words that we run into trouble.
There’s an economic reason to be worried about AI too.
Jobs will be lost because of this advancement in technology. Fewer writers will be needed in corporate environments, and a host of administrative positions will be eliminated. We haven’t even touched on how AI technologies like Midjourney and DALL-E can generate highly complex images with a simple text prompt, cutting out the need for many concept artists and graphic designers. It should go without saying that if businesses can cut costs by replacing workers with technology, they will. What will replace those jobs? Editors will be in demand, certainly, but one editor can supervise a dozen writers. One designer can cull through hundreds or thousands of AI-generated images. No one has done the work to ensure we adopt this technology responsibly without a significant impact on our workforce.
AI came for the creatives first, but it will not stop there.
ChatGPT is already capable of writing code more quickly and efficiently than most beginner and intermediate programmers. “Learn to code,” became the most common response to people who were losing their factory jobs the last time automation sparked massive layoffs. That seems like even worse advice now.
In a perfect world, governments would put policies into place to ensure that human labor always takes priority over AI and automation. But we’ve seen nothing like that to date. Many free-market capitalists would likely protest such a move. Even in that perfect world, government action would only be able to protect our livelihoods. No policy can address our common loss of imagination, emotion, and empathy, or our diminishing capacity to communicate with each other as fellow human beings.
I don’t have an answer to this problem, but a good place to start is by having conversations like these.
Words are how we think, how we feel, and how we relate to each other. They’re far more important than we give them credit for. By outsourcing our thinking and our emotions—our words—to AI, I believe we will lose far more than we gain.