We apologize for the inconvenience caused by text appearing broken or incorrect when generating images with AI.
1. Cause: Technical limitations of AI
Current AI image generation technology has common limitations in font recognition and processing. This may result in imperfect output when generating images containing text.
2. Solutions
Here are two methods to work around AI’s technical limitations and get better results.
Method 1: Specify a model with good text performance
Currently, the ‘ChatGPT Image model’ has the best text output performance. Try explicitly requesting AI as follows.
Example prompt: “(Enter desired content) Use the ChatGPT image model”
Method 2: Generate without text, then post-edit
The most reliable method is to generate only the source image without text using AI, then add text directly using professional editing programs like Photoshop or Canva.
We are continuously introducing the latest AI models to improve text processing performance. We apologize again for the inconvenience caused by AI technology limitations, and hope you can achieve better results using the methods above. 
