In just two years, the use of generative AI has gone from taboo to tolerated to acceptable. In some cases, it’s even expected. 

But there’s one group of people that has zero tolerance for lazy AI writing: your customers.

Without a human editor, AI content is bland and uninspired. It’s generic. It’s — dare we say it — bad. When it doesn’t get the TLC it needs, the consequences can be dire: students get flunked, lawyers get sanctioned, and brands lose the trust and interest of their customers.

Thankfully, there are a handful of things you can do to take your AI content from generic to genuine. Here’s what you need to know.

The Problem with AI Content

At a glance, AI writing may look good. You might even think it looks professional, complete with some five-dollar words that make you look smart. But the problem is that AI can’t think or reason like a human can. It’s incapable of conjuring an original thought. Even its style and voice are recycled from other writers and then mashed together. As a result, it just doesn’t hold up to close examination.

This means that you can’t afford to simply create AI content and hit Publish. You may think you’re saving money, but today’s consumers have finely tuned BS detectors and can spot disingenuous content from a mile away. If they get one whiff of lazy AI content, they’re either scrolling past, calling you out on it, or hitting Unsubscribe. All this translates to significant lost revenue.

But all hope is not lost! Now that you know why AI content falls short, you can start to improve it. 

6 Tips to Make Your Copy Sound Human

So what’s the secret? How do you clean up (and spice up) your AI-generated content and make it publish-ready?

Here are six tips for doing just that.

1. Use Clear Prompts to Set the Stage

First, let’s work on improving what you have to work with. After all, it’s far more difficult to transform completely generic, vague content into something remotely acceptable. So if you can use a clear prompt and get better output to begin with, you’ll start off stronger.

Consider this example, straight from ChatGPT:

Prompt: “Give me a one-sentence value proposition for my tech startup that creates translation software.”

Output: “Empowering global communication through seamless, AI-driven translation software that bridges vernacular barriers in real time.”

This is a good example of one of AI’s biggest weaknesses: when it doesn’t have enough detail to go off of, and when you don’t give it a specific voice to use, it quickly falls back on nausea-inducing buzzwords and corporate-speak.

Let’s try again.

Prompt: “Give me a one-sentence value proposition for my tech startup that creates translation software. It’s compatible with any phone/earbud combo, and it translates foreign speech and plays it to you while the speaker is talking. Then you say your response, and your phone will either translate it in real time or tell you how to say it in your ear. The target audience is international students. Our mission statement is to empower expats to live the life of their dreams wherever they are in the world and help them build genuine relationships with the locals. The tone of the value prop should be fun, warm, and personable, without jargon or complicated language.”

Output: “Helping international students connect and thrive abroad with real-time translations that work with any phone and earbuds, so you can chat naturally and build friendships no matter where you are.”

It still has some obvious problems, but it’s much closer now. This is a result you can work with and tweak, whereas the previous version was bordering on meaningless.

2. Play Word Doctor to Soften Robotic Phrasing

How many times have you seen the word “delve” in recent history? How about the phrase “Let’s embark on a journey through the rich tapestry of X”? 

Chances are you’ve seen language like this more times in the last two years than you have your entire life. There’s nothing inherently wrong with it, but when AI tends to over-rely on the same vocabulary, the result is samey, robotic content that’s painfully boring to read.

3. Bring a Human Touch by Adding Personal Experience

People don’t connect with faceless machines; they connect with stories. Whether it’s weaving anecdotes into your company blog or focusing on your own personal insights in a LinkedIn post, stories are what drive engagement.

This is something you’ll have to do yourself. AI doesn’t have your life experiences, and even if you share them with the tool, it consistently fails to understand their relevance. This means you’ll need to find appropriate areas within your content to incorporate your genuine story. It’ll take some time to get right, but it’s well worth the effort — this is what builds lasting customer relationships.

4. Break Up Walls of Text by Adding Visuals

Visuals are great for catching the eye and keeping readers engaged. If nothing else, AI is great at creating a wall of text, which can quickly feel overwhelming to the reader. An aesthetically pleasing image in the right spot can break up the flow and keep things interesting.

If you use AI to create your images as well, make sure you carefully review them. Too many fingers on someone’s hand or a jumbled mess of letters on a sign in the background are telltale signs of careless AI use, and it’s just another way to make your reader jump ship.

5. Keep It Short and Simple

AI loves its long, complicated passages. And when its text regularly contains words like “innovate,” “empower,” and “elevate,” it’s unsurprising that AI content can be quite boring to read. Instead, take a page from Shakespeare, who famously said that “brevity is the soul of wit.” Use shorter sentences, vary your paragraph length, and eliminate fluff.

You’d be surprised just how powerful this can be.

In our earlier example, consider the second version of our translation software’s value prop: “Helping international students connect and thrive abroad with real-time translations that work with any phone and earbuds, so you can chat naturally and build friendships no matter where you are.”

Let’s trim the fluff and make it tighter, shall we?

Result: “We help students chat naturally, build relationships, and thrive around the world with real-time translation on any device.”

It’s not perfect by any means, but it’s come a long way and is much closer to where it needs to be.

6. Thoroughly Fact-Check

Whether it’s creating content based on outdated information or flat-out making things up because the words “I don’t know” are not in its vocabulary, artificial intelligence is downright notorious for its unreliability. So if you don’t carefully examine your content and fact-check it where needed, you could inevitably damage your credibility beyond repair.

Don’t be afraid to ask it where it got certain information. You’d be shocked at how often you see the words “You’re right. That doesn’t appear to be the case. Let me rephrase that.” And even if it does tell you where it got its information, it’s crucial that you check it out for yourself. Your brand’s reputation is at stake, so don’t skip this step.

Find the Right Partner for Your Brand

AI is not a replacement for your content strategy — it’s your partner. You have to work with tools like ChatGPT and not trust them to do all the work. But chances are that you’ve been using AI to avoid all this extra work. Why use it at all if you still have to spend hours making your content shine?

That’s where Presscart comes in. We’re a full-service partner that can take your brand’s mission and turn it into high-quality content that drives engagement and conversion. We use AI ethically and responsibly, enriching our workflow without letting it compromise our integrity and quality.

For high-impact stories that connect with your audience, contact our team today to get started.