ContentAI currently tries to generate an image (with or without content title, CTA, etc.) for placing content in an email. However, this results in the predictive content failing to have much customizability in display, and more importantly, causes the predictive content to be inaccessible. The final code output is an image with no alt text or identifying title in the link.
Rather than trying to dynamically generate an image with all content included, could the setup be changed to retrieve the image, URL, title, and call to action as separate items? This both prevents having to create more image template styles or customization for customers and allows developers to code to standards for accessibility compliance.
.png)