Artificial intelligence has opened up amazing new possibilities for designers to create incredible graphics and images. In this post, I’ll share how I leveraged two of the hottest AI art generators – Dall-E 3 and Ideogram AI – in an epic showdown to create amazing t-shirt designs.
Over the past year, AI image generation tools like DALL-E and Midjourney have exploded in popularity. They allow you to generate nearly any image just by entering a text description. Recently, Anthropic released Dall-E 3, taking this technology even further.
Another newcomer is Ideogram AI, which also creates unique vector images from text prompts. I decided to test these two AI tools head-to-head to see which could produce the best t-shirt design.
First, I gave each AI the prompt: “Red lollipop vector image with text saying ‘Do you want to lick this?'”
Dall-E 3 rendered an incredibly detailed lollipop complete with 3D shading. The text wrapped around the lollipop was clear and properly spelled.
Ideogram AI also generated a cool vector lollipop design. However, the text had some spelling and formatting issues.
Next, I wanted to create a frog design. I searched Merch Informer for frog shirt ideas for inspiration. Then I gave each AI this prompt:
“Green frog vector image with text saying ‘Frog Dad'”
This time, Ideogram AI created a more polished, clean vector design of a cute frog and custom text. Dall-E 3 had some artifacts and glitches around the text.
However, I was able to take the Dall-E 3 image and upscale it using Vectorizer AI for a crisper final image.
For the finishing touches, I imported the AI generated designs into Photopea and Canva to remove backgrounds and perfect my t-shirt graphics.
The results prove that both Dall-E 3 and Ideogram AI are amazing tools for rapidly creating custom vector designs. While Dall-E 3 generated more photorealistic images, Ideogram excelled at clean vector graphics. With a bit of post-processing, the outputs from both can become print-ready designs.
By combining these AI art platforms with upscaling tools and graphic editors, designers can skip over the tedious illustration process and simply generate on-demand graphics with these AI assistants!
I had a blast pitting these two creative AIs against each other. It’s fascinating to see how rapidly artificial intelligence is augmenting and enhancing the design process. The future possibilities seem endless.
Have you experimented with any AI art generators? Let me know if you have any questions in the comments!
RJ Martinez 