A client walks in with a foamboard job. They've generated a gorgeous image using an AI tool — vivid colours, sharp-looking details, exactly what they wanted. They send it over and ask for it to be printed at 4×8 feet. The printer opens the file. It's 1024×1024 pixels. What happens next is a conversation that no one enjoys — and it's happening in print shops across Malaysia every single week.

The Foamboard Story — What Actually Happened

Let's start with a real situation. A client needed a 4×8 feet foamboard print — the kind you see at exhibitions, events, and retail displays. Instead of hiring a designer, they used an AI image generator. The result on their phone screen looked stunning. Crisp, colourful, professional.

They sent the file over. The printer opened it and immediately saw the problem: the image was generated at 1024×1024 pixels. At 4×8 feet, that works out to roughly 21 dpi. For context, a good large format print needs at least 100–150 dpi at final size, and ideally 300 dpi for anything viewed up close like a foamboard display.

The printer explained this. The client's response? "Just redraw it for me. For free."

And here's where it gets interesting — because this is not an unreasonable thing for a client to feel. They genuinely didn't know. They saw a sharp image on their screen and assumed it would print the same way. The misunderstanding isn't about bad faith. It's about a gap in knowledge that AI tools have made dramatically wider in the past two years.

The printer, for the record, was not being difficult. Redrawing an AI-generated image at print resolution is skilled work that takes time. It is a paid service — starting from RM150 depending on complexity — not something that comes bundled with the printing cost. More on who owns this problem shortly. First, let's understand why it happens at all.

Resolution Explained in Plain Language

Resolution is just a measure of how much detail is packed into an image. It's expressed in dpi — dots per inch — which tells you how many individual colour dots exist in every inch of the printed output.

The higher the dpi, the more detail, the sharper the print. The lower the dpi, the fewer dots, the blurrier everything looks — especially at large sizes where each dot gets physically bigger.

When you generate an image using an AI tool like Midjourney, ChatGPT's image generator, or similar platforms, the output is typically a fixed pixel dimension — 1024×1024, 1792×1024, or similar. Those pixel dimensions are designed for screen use. They look excellent on a monitor. They are not designed with print dimensions or dpi in mind, because the AI has no idea you plan to print on a 4×8 feet foamboard in Kota Kinabalu.

The Math That Breaks AI Images at Large Format

This is the part nobody shows you — and once you see it, you'll never look at an AI image the same way again.

Let's take a typical AI output: 1024 × 1024 pixels.

At 300 dpi — the standard we print at — that image can be printed sharply at a maximum of:

1024 ÷ 300 = 3.41 inches × 3.41 inches

That's roughly the size of a large postage stamp. Print it any bigger and quality degrades.

The numbers don't lie. An AI image that looks flawless on a 6-inch phone screen becomes an embarrassment on a 4-foot foamboard. The pixels haven't changed — they've just been stretched across a much larger physical area, and suddenly every gap between them is visible.

Can't You Just Upscale It?

This is the first thing clients ask. And it's a fair question — after all, there are tools that claim to make images bigger. The answer is: sometimes, partially, within limits, and someone still needs to do the work.

AI upscaling tools like Topaz Gigapixel, Adobe Firefly's upscaler, and Magnific AI use machine learning to intelligently add detail when enlarging an image. They are genuinely impressive and far better than the old "resize and blur" method. However:

• They work best when starting from a reasonably high base resolution. Starting from 1024px and trying to reach 4×8 feet at 300dpi requires a roughly 35× enlargement — beyond what any upscaler handles cleanly.
• Results vary heavily depending on image content. Faces, text, and fine details often come out wrong — distorted features, garbled letters, smeared edges.
• The process takes time and skill to do properly. Someone needs to run the software, evaluate the output, make corrections, and prepare the final print-ready file.
• That someone is doing paid professional work — it does not come free with your print order.

Who Owns the Problem?

Here's the part of the conversation that nobody likes but everybody needs to hear.

When you bring a file to a print shop, the printer's job is to print it. If the file is unsuitable for print — wrong resolution, wrong colour mode, missing bleed — that is a prepress issue, and it sits with whoever supplied the artwork. That is standard practice across the entire printing industry, everywhere in the world.

The printer did not generate your AI image. The printer did not choose the output resolution. The printer did not approve the file for large format use. Asking a printer to fix an artwork problem for free is like bringing a car with a flat tyre to a mechanic and asking them to fix it as part of the parking fee.

If a printer tells you your file has a resolution problem, they are doing you a favour. The alternative — printing it anyway and letting you discover the blurry result after the fact — would be far worse.

When AI Images Actually Work for Print

To be fair — AI image generation is a genuinely useful tool. The problem isn't AI. The problem is using a screen tool for a print job without understanding the difference. Here's an honest breakdown of where AI images work and where they don't:

The rule of thumb: if it's going to be larger than A3 and viewed closer than 2 metres, an AI-generated image is almost certainly going to be a problem. If it's small format or purely digital, AI images are perfectly fine — and often excellent.

The Smarter Alternative — Canva Done Right

If you don't have a graphic designer but still need print-ready artwork, Canva is genuinely a good option — if you use it correctly. Most people don't, and that's where the problems start.

Here's exactly how to get a Canva design that's ready for print:

What Graphic Designers Do That AI Still Can't

There's a lot of talk about AI replacing graphic designers. For certain tasks — generating concept images, creating social media visuals, making quick mockups — AI tools are genuinely competitive. But for print-ready work, a graphic designer is doing something fundamentally different that AI does not do.

AI image generators are creative tools. Graphic designers — especially ones familiar with print — are technical production professionals. The two are solving very different problems. A designer who understands print knows about CMYK colour profiles, bleed areas, safe zones, and file formats. An AI image generator produces a JPEG and calls it done.

Will AI eventually replace print-ready graphic design? Perhaps, one day. Today in 2026, it absolutely has not. The gap between "looks good on screen" and "ready to print at 4×8 feet" is still a human gap — filled by skill, experience, and understanding of the physical output.

The Print-Ready File Checklist

Before you send any artwork to a printer — AI-generated, Canva-made, or professionally designed — run through this quick checklist. It will save you time, money, and that awkward conversation.