Hey, author here, I'll attempt to clarify.

In a data grid or table the relative cognitive load of the page is already very high. Adding iconography to the table body content is often unnecessary and increases visual noise, processing requirements, and generally reduces readability/scanability.

I've always felt that icons in this context are a risk or liability instead of a strength. I decided to info dump my findings to my team then published it as an article.

I probably could use a good editor to help me next time!

I think you need to show examples of this bad usage, then.

Your first image has zero icons in the rows. It has album covers but those aren't icons.

Your second and third images show very usage that combines text, color, and relatively standard icons like checkmarks, X's, or in-progress. These are good and if you're trying to suggest these reduce readability, scannability, or add noise, then I'm frankly baffled.

Your third image also shows profile images, but again those are not icons.

So what are you arguing against? I can't ever remember coming across icons in a datagrid that "added to the relative cognitive load". And if you're arguing against checkmarks or X's, I don't think your arguments hold up.

But even with "real" icons -- like, I've seen icons to show if a software package is for Windows or Mac or Linux. If a row is a TV show or a movie. If it's one file or an archive. If there's a PDF file download attached. An alarm icon for something past due. But these all seem totally fine and helpful. They're generally linked to a major feature of the platform that everybody understands, and help scannability.

Without clear examples of what you're arguing against, I'm frankly completely lost.

In-line images should almost always include a caption, especially in a "do this and not that" type of piece