AI/ML, Pixels, Energy & Schemas

Recently, at a lifelong learning session at a local retirement home, the speaker said AI/ML applications are so energy intensive because they analyze every pixel.  While that may be understandable for images, I found it hard to believe with respect to textual documents.
In response to my chat with it, here's what Gemini concludes with respect to the potential of such applications to take advantage of internationally standardized XSDs:

Overall, while LLMs don't yet fully utilize the semantics and structure of schemas, the field is actively exploring innovative ways to integrate this information for improved performance in various tasks involving structured textual data.

In the meantime, their failure to do so might be taken as an example of artificial ignorance as well as a tremendous waste of time and energy that might be more productively spent.
Owen Amburhttps://www.linkedin.com/in/owenambur/