Designing with AI

Designing with AI: useful tool or risky shortcut

Artificial Intelligence (AI) is everywhere right now but what does it mean in the context of design? 

At its core, AI in design usually refers to generative AI: systems trained on vast datasets that can produce original-looking outputs, from images and videos to layouts and even text prompts. Tools like Midjourney, DALL·E, and Adobe Firefly let users type in a phrase and receive fully formed visuals in seconds. These platforms have changed the creative process, offering designers endless possibilities for ideation and execution. 

The integration of AI into creative workflows is not just a trend but a significant shift. According to McKinsey’s 2024 survey, the adoption of generative AI in business functions surged from 33% in 2023 to 71% in 2024, with marketing and sales among the top areas of implementation. This shows a broadening acceptance and reliance on AI tools across industries. 

We’re seeing both sides of AI at The Graphic Design House. AI can be a great time-saver and idea-starter, but it also raises big questions about originality and ownership. Like most designers, we’re excited by the possibilities, but we’re keeping a keen eye on where and how it’s useful. 

What AI in the design industry looks like 

AI in design isn’t a futuristic concept anymore, it’s already showing up in day-to-day work, though usually in very specific ways. 

We’re seeing generative AI most commonly used during early-stage ideation, generating quick mood boards or visual prompts that help spark ideas. These can help with clients who struggle to visualise an idea and shows the direction for how campaigns look and feel. All this can be done quickly and efficiently with a handful of creative prompts (which is a skill in itself). It’s not about finished work; it’s about getting the conversation started faster. 

Another area where AI is creeping into daily workflows is background removal and image enhancement. Tools like Adobe’s Firefly have been embraced by a lot of creatives because they’re built into tools we already use, and they save time on the fiddly stuff. 

But it’s important to say: the quality varies. You can get weird objects, inconsistent lighting, or people with way too many fingers (or not enough fingers!).  

We’ve found that using AI tools effectively still requires a designer’s eye, to pick the best outputs, to tweak them, or to know when it’s better to bin it and start from scratch. Nobody’s handing over AI-generated visuals as final artwork (and if they are, there are probably issues, more on that in the next section). 

The problem with AI in design 

AI has its uses, but it also brings a lot of serious challenges, especially when you move beyond ideation and into commercial or client-facing work. At The Graphic Design House, we’re already seeing the complications AI can cause, and in many cases, it’s just not practical (or ethical) to use without caution.

1. Legal grey areas, who owns what?

One of the biggest concerns with AI-generated art is copyright and ownership. Most generative AI tools are trained on huge data sets scraped from the internet, often without the consent of the original artists. That means the output might look new, but it’s built on the backs of uncredited creatives. 

In some cases, artists have found their work directly mimicked by AI, like fantasy illustrator Greg Rutkowski, whose name had been used more than 400,000 times as a prompt in AI tools since September 2022. 

There are also legal battles underway. Getty Images filed a lawsuit against Stability AI in early 2023, claiming the company used millions of copyrighted images to train its model without permission. These cases are ongoing, but they highlight the murky legal territory we’re in. If you can’t prove the origin of your image, you can’t guarantee it’s safe to use commercially.

2. Not fit for print

AI-generated visuals might look impressive on screen but that doesn’t mean they can be used in real-world formats like print. 

Most images generated by tools like DALL·E or Midjourney are low resolution, usually around 1024 x 1024 pixels at 72dpi (Dots Per Inch). That’s fine for web or internal mock-ups, but completely unusable for print where it’s generally recommended to be at 300dpi. This means that if you’re taking content straight from generative AI it’s highly unlikely you will be able to produce high quality printed materials.

3. Lack of originality

Even when AI output looks good, it can feel… generic. That’s partly because these tools rely on existing styles and trends, so a lot of what comes out tends to resemble what’s already popular or trending at the time. 

There’s also a noticeable “AI look”. Overly smooth textures, strange lighting, warped body parts (especially hands and feet), inconsistent text. It can stand out in the wrong way and lead to trust with prospects and customers decreasing. 

And then there’s the question of design intent. Good design solves problems, tells a story, and connects to context. AI doesn’t understand strategy when designing it just knows what looks good based on past data. That might work for a concept sketch, but it doesn’t replace the thinking behind a finished piece.

4. Ethics and Trust

Using AI without transparency can damage trust, especially in industries where integrity and originality matter (think: theatre, education, non-profits). Clients want to know what they’re paying for. If they find out later that part of a campaign image was generated by AI, it can raise awkward questions about originality, IP rights, or simply value for money. 

We’ve also seen that team morale can take a hit when AI is overused. Designers want to create, not curate outputs from a machine. There’s a difference between using a tool to accelerate your work and feeling like your craft is being replaced. 

Tools not replacements

AI has found a place in the design industry but it’s not replacing designers any time soon. Right now, the most valuable use of AI is as a support tool: for speeding up repetitive tasks, helping with early-stage ideation, or mocking up visuals quickly. Beyond that, the limitations become clear legally, ethically, creatively, and technically. 

At The Graphic Design House, we’re open to using AI when it adds value but we’re cautious about how and where it’s used. The integrity of our work matters. So does originality. And ultimately, the best creative still comes from human insight, not just machine output. 

AI can spark ideas. It can speed things up. But it can’t replace what makes good design actually work: thought, context, and connection. 

Our work in action