AI is not the problem. Lazy, generic and unedited AI design is.
That sounds like a win. But there is a growing problem.
The internet is filling up with what many people now call “AI slop”, large amounts of AI-generated content that may look acceptable at first glance but often feels generic, repetitive or disconnected from the brand behind it.
For businesses, that matters.
What is AI Slop?
You have probably seen it online.
It might be an obviously artificial image with strange hands or unrealistic details. It might be a social media graphic that looks almost identical to hundreds of others. It could be an article, advertisement or image produced quickly with little editing or thought behind it.
The problem is not necessarily that AI was used. The problem is using it without taste, experience, strategy or human judgement.
AI can produce something in seconds. That does not mean the result is useful, accurate or appropriate.
The same applies to AI graphic design.
A business can generate a logo, advertisement or social media image without understanding whether it actually represents the business. The result might look impressive on a screen but fail to communicate anything meaningful.
AI Graphic Design Is a Tool, Not a Design Strategy
I recently had a client experience that probably explains this better than any theory.
The client needed a new logo, so I created five different concepts exploring different visual directions. Unfortunately, he didn’t like any of them, so we went back to the drawing board.
Fair enough. Design is subjective.
I suggested he look at some Google Images and show me examples of logos he liked so I could better understand the direction he wanted. Instead, he sent me a logo he had created using AI.
He preferred it.
Again, that’s fine. AI can produce some impressive visuals.
But then I asked a few basic questions. What colours were being used? What fonts? Could I get the logo in vector format so it could be properly used across the website?
The response was essentially, “What’s RGB?” and “What’s a vector?”
And that’s where the difference between creating a picture and creating a brand becomes important.
RGB, fonts and vector files aren’t designers trying to make things complicated. They’re basic parts of making sure a visual identity works properly across a website, signage, print, vehicles and other applications.
The AI had created a logo. It hadn’t created the brand system around it. That’s the part that still requires human judgement and experience.
The funny thing is, if he spent as much time telling me the design he wanted as his son spent prompting AI, we would have designed something much better and it wouldn’t look pixelated and like it was half broken like this.
He also changed his business name as the AI came up with something different.
(i definatley am not holding a grudge. no not at all, if you want more information about this, lets meet up for a tea and I will spill it!)
Good Graphic Design Starts With Understanding the Business
The AI had produced an image. It hadn’t considered the website or whether the logo would even work in different formats.
And that’s not really a criticism of AI. That’s simply not what it was asked to do.
The prompt was designed to create an image. My job as the designer was to think about what happens after the image exists.
Someone without a strong understanding of graphic design can look at an AI-generated image and see something that looks good without necessarily recognising the nuances that make a design effective, practical and consistent.
AI can produce the starting point. The knowledge to judge whether that starting point is actually right comes from experience.
That’s the part that can easily get lost when we talk about AI graphic design.
But Couldn’t You Have Just Used AI?
I’ve been designing websites and graphics since 1999, so I’ve seen the industry change dramatically. I’ve gone from designing things manually to desktop publishing, Photoshop, digital photography, stock libraries, online design tools and now generative AI.
I’m not against AI. Quite the opposite.
But there’s an important difference between using AI to create something and knowing whether what it creates is actually good.
AI can give me 20 possible directions in seconds. My experience helps me work out which three are worth exploring. It can generate an interesting logo, but experience tells me what questions to ask. Will it work at different sizes? Can it be reproduced properly in print? Does it make sense for the business? Is it distinctive enough to build a brand around?
That’s where experience matters.
AI can make the design process faster, but the final judgement still comes from years of actually doing the work.
AI Can Make Everything Look the Same
A logo doesn’t need to explain everything your business does. In fact, trying to squeeze your entire business into a logo can often make it harder to recognise and remember. A good logo gives the brand something distinctive to build around. The rest of the visual identity does the explaining.
This is where AI can struggle.
It might look at a design and decide that a pixelated edge is a shadow, or interpret a visual effect completely differently from what the designer intended. It can also get typography wrong. A font might look almost right, but the spacing, proportions, weight or letterforms can change the whole feel of a logo.
Those details matter.
Design Needs to Work, Not Just Look Good
A designer isn’t just asking, “Does this look good?”
They’re asking whether the logo will still work when it’s printed small, placed on a website, embroidered on a shirt, used on a vehicle or incorporated into digital marketing campaigns. They’re also thinking about how the colours, fonts, imagery and layout work together so the business looks like the same business everywhere.
This is what AI fails to grasp
When the logo, fonts, colours and other visual elements start looking inconsistent across different platforms, the business can quickly lose some of the recognition and trust it has worked to build.
AI is a useful graphic design tool . But knowing what should be designed, what should be changed and what should be left alone still takes human judgement.
Good Design Still Needs a Human
I’ve spent years learning how design works, so when my expertise is questioned simply because AI can generate a nice-looking image in a few seconds, it’s a little amusing. The tools may have changed, but good design still requires experience, judgement and knowing what you’re actually looking at. I think it’s fair to point out that there’s a little more to design than pressing a button.
You still need to understand the business. You still need to understand the customer. You still need to make decisions about colour, typography, layout, messaging and consistency. You still need to know where a design is going to be used.
And sometimes you need to tell a client that the thing they think looks amazing isn’t actually going to work.
That’s not because designers are trying to protect their jobs from AI.
It’s because that’s the job.
AI is a useful tool. AI slop is not.