Please can we have Chromium’s image description feature back? It lets a blind person (like me) gain some insite into an image that is otherwise without description. It’s not much, but it’s always better than nothing. In Chrome, the feature is found in Settings, under Accessibility. You could make it more privacy-preserving, by proxying the requests, or using your own (or an alternative) cloud-based image and/or OCR service. In any event the feature should, like Chrome’s, be configurable so it can be activated on-demand or always at user’s discretion.
There are still faint outlines of the feature in Brave, e.g. in image context menus. But AFAICT it doesn’t work once activated.