If you have used Generative AI tools to create images, you may have encountered problems when writing text in images. Knowing why this happens will help avoid this problem, as well as find a solution.

Why can’t AI write text in images?

The biggest reason why AI cannot write text is due to past input data. While AI is quite good at generating images in general, some aspects such as text have not been used at the same level to train large language models (LLM). Therefore, AI does not have much skill in creating text in images.

Although already quite advanced, AI tools are technically still in their early stages. Furthermore, some AI tools are running out of training data. It’s still possible that generation tools can generate better text, but for now, you’ll have to try an alternative.

Are there any workarounds for unreadable text in AI images?

Although creating text in AI images is quite difficult, it is not impossible. Here are the most effective methods at the present time…

1. Give the tool more specific prompts

When first starting to use AI, people’s prompt writing skills are often not up to standard. The biggest mistake is giving vague prompts like “create an image of a street scene”.

No wonder you get substandard results. The best way to use Generative AI is to think of your prompt as a creative brief. The more specific, the better. Here is an example prompt:

Create an image of the exterior of an Italian cafe. I want it to have the writing “Café” as a sign, and it should be on a sunny day.

Roughly translated: Create an exterior image of an Italian cafe. I wanted the image to have the words “Café” on the sign and on a sunny day.

As you can see from the image below, ChatGPT did quite well. This wouldn’t happen if you just said “Create the exterior of an Italian cafe”.

From personal experience, simple images seem to be more effective. For example, the image above has fewer elements than the image tested below – meaning the AI ​​has less to try and learn.

2. Use variations of the word “Text”

Changing the language in the prompt a little can make a big difference. After trying the term “text” several times, you’ll start to feel like you’re going in circles, so try different synonyms.

For example, instead of “text”, you can try:

• Title
• Alphabet
• Written word
• Symbol

If none of these work, keep trying alternatives. What succeeds and what doesn’t can vary depending on what you’re trying to create. For example, you probably wouldn’t use “symbol” if you were trying to design a birthday card. Experimenting with different AI prompt ideas is key here, and different tools can work in different ways.

3. Add text using replacement tools

You shouldn’t add text through alt tools if the words are supposed to be part of the image, but it’s a good idea if you’re trying to create more graphics. It will work if you want to design a card, graphic, book cover or something similar.

If you are going to do this, make sure there is enough space in the image. You can use any tool you want to add text to images. Canva or Adobe Photoshop Express are recommended options, but other apps that let you add text to images on your smartphone will also work with content you create with AI.

You should also make sure that the text matches the AI-generated image. Use fonts that match the mood you’re trying to convey and reshape things however you feel is necessary.

4. Keep the text short

Creating text in AI images that fail is usually because you try to add too much text. Honestly, this happens whenever you try to create anything with more than 10 characters. Similar to the suggestion to keep images simple, you should do everything possible to keep your text concise.

To test this, the author tried another prompt:

Please generate an image of a bank with the text “bank” on the outside. I want it to be in a downtown city area and a modern structure, like what you would see in a US city.

Roughly translated: Let’s create an image of a bank with the word “bank” on the outside. I wanted it to be in the downtown area and have a modern structure, like what you would see in a US city.

As you can see below, the AI ​​engine did a pretty good job of listening to users. However, there are some errors in part of the image. So tell the AI ​​you only want one or two signs to limit the chances of this happening. These problems also seem to occur when text is smaller, so keep that in mind.

5. Use tools to correct text errors

Similar to how you can remove errors from photos taken with a camera, you can try using a tool that helps fix text errors in photos. There are many applications that fix nonsense text in photos created by AI, such as Storia Lab.

When using these tools, you can typically select text and change the content accordingly. Some apps are free, but in other cases you’ll need a paid subscription. If you create a lot of images using AI, you can purchase a monthly or annual plan to access it whenever you need it.

6. Use an AI engine that generates accurate text

You may have heard of Midjourney, DALL-E, Firefly and similar AI image generators. But did you know that there are specific AI artwork creation tools that create exact text in images?

Since one of Ideogram’s specialized features is creating easy-to-read text, you can definitely try this app out for some of your creations. Ideogram has a Magic Prompt feature that extends your original functionality, meaning you can get even more accurate results. In addition to the free plan, paid subscriptions start at $8/month.

While the resulting AI has shortcomings in creating images with readable text, you’ll find a few alternative solutions if you’re willing to think outside the box. Sometimes the best option is to give your tool detailed prompts with fewer words to process. You may also want to fix problems in your images with other software or use tools specifically designed to create accurate text.