You’ve heard all about AI by now. It’s impossible to avoid. These generative programs that sweep any available internet resources and respond "intelligently" are changing the digital landscape. The meteoric rise of AI has made it clear: Get with it or get out of the way.
What does this mean for content? AI solves the problem of creating content, doesn’t it? It basically takes care of everything, right? Well… Not quite. Let’s look at some of the drawbacks we’ve encountered when using AI for content generation.
First Thing’s First…
AI isn’t all bad! There are plenty of circumstances in which AI improves efficiency and boosts productivity. Members of our team have fed ChatGPT spreadsheet data to see which tasks take the most time to complete, and they’ve then asked it for recommendations to enhance the team's workflow based on that data. Super helpful! We’ve also used it to summarize large pieces of information like blog posts–inputting the full text and then having it broken down into succinct paragraphs. When you’re synthesizing data, it’s a lifesaver, and saves a ton of time! Try doing any of that manually. It could take hours.
All that is to say that we’re not anti-AI. We’re simply proponents of using these tools responsibly and ethically, as you should be. There are aspects of it that you’d do well to be aware of from a creative standpoint, however…
No One Knows Your Brand Like You
Your brand voice is unique to you and speaks to your partners and audience in a way that they’ll find familiar and comforting, and you may think that AI is smart enough to replicate this specific way of speaking.
It can come kind of close, but it’s hard to get it to be spot on. When it comes to recreating your voice, it is a matter of trial and error regarding the prompts you’re inputting into ChatGPT. For us, it took quite a bit of experimentation, and it still required that we rewrite the content that the program was putting out.
The Viability of the Art
When it comes to visual assets, AI generators such as Midjourney or Dall-E can take prompts and generate an image that meets your needs, but there are several issues to be aware of when considering this.
Inhuman Creation Station
The first is that people can tell when an image is AI generated–especially if you’re using it to try and make photorealistic human beings. While AI may have mostly fixed its issues with teeth and fingers (if you don’t know what we mean, look it up), it still created pictures that are vaguely… unnatural. It has an uncanny valley effect where the observer knows they’re looking at something that isn’t human, which isn’t a good look for you. Need pictures of people? Go for stock images. They may not be exactly what you’re looking for, but they won’t give people the creeps.
The Stigma of the Source
Even if you’re not trying to generate photorealistic people, you could still run into issues with other graphics you’re generating. Even basic graphics can show telltale signs of AI usage, which can rub your partners and prospects the wrong way.
Many artists have come out against the use of AI for image generation because it’s sourcing their art without giving them any credit or compensation. Many people simply will not use it for this reason, and there’s significant pushback regarding it. This is a well-known issue, and as a result, the use of AI art in your assets and marketing can come across as cheap and predatory. People can read it as your organization not wanting to pay artists, and that can turn a lot of people away. Not everyone, necessarily, but enough that you might want to consider it.
Very Confident Misinformation
ChatGPT will write you a step-by-step tutorial for anything and everything you want to know… Now will all those steps be accurate and make sense? Well…
We’ve found that it doesn’t give the right steps, tool names, or anything you would need for a successful tutorial. Remember, Artificial Intellligence isn’t actually intelligent. It’s just flopping around the internet finding relevant words and stringing them together in a way that makes sense to a mindless algorithm. The last time we had it generate a tutorial, the entire thing had to be rewritten from scratch
If you’re going to use AI to write a blog post for your knowledge base, you need to be prepared to fact-check every single aspect of it, or you risk spreading misinformation and losing credibility.
Professionals End Up Undervalued
AI is shaking up the business landscape in many ways, but some of those ways are harmful to the creatives that have helped you get to this point. Because ChatGPT and Midjourney are being used to replace the vital roles of copywriters and graphic designers, it’s changing how those positions are valued, while they still have the same level of importance.
You still need copywriters to fix ChatGPT outputs, and you still need graphic designers to modify art outputs, making sure these graphics abide by best practices and don’t put people off. However, some organizations are using the generation aspect of these tools to rewrite the job descriptions of these experts, using it to justify paying them less. They become AI minders instead of artists, which hurts everyone, but especially them.
