Adrian Roselli made a lengthy post in which he feeds self-selected images into CLIP interrogator, an AI tool for describing images, and then feeds this text into Stable Diffusion. The same he also does with his self-provided alt-texts. By hiding the images he allows you to make this exercise using your own imagination as well (and sighted users can then compare the results). His conclusion:
(…) Sadly, these automated prompt generators themselves rely on context many users may not have, such as naming artists or styles.
At the very least, these prompts would need to be edited for human consumption. The art sites, artists, uses, and strings of numbers are meaningless on their own. Simply removing them does nothing to make the prompt itself useful as alternative text.
In short, this is also a flawed approach and no organization should consider it a viable option.