is auto-tune just promoting poor singing and laziness on the part of the engineer?
Good question. I think the answer is "no." From my experience (and I'd be curious to hear what other engineers say), singers who experience their own voice being pitch-corrected usually end up being better singers for it.
If I'm working with the same singer over the course of an album or several albums, the rule, and not the exception, is that they become more accurate (with pitch and time) as time goes by.
For one thing, hearing yourself singing out of tune on playback, seeing a graph of just how far out of tune you sang, and then hearing what you should
have sung (once it's been corrected) is extremely humbling and motivating. For a lot of people, this is the first time they even realize that they actually do
sing out of tune, even if they've been told so by their MD 100 times.
Also (and this part is a little more out-there) hearing yourself sing something in tune, even if you didn't actually do it, acts as a sort of creative visualization.
As for the actual recordings, now that pitch correction (as a corrective tool, not an effect) is a fact of life, getting too focused on pitch during a recording session is only a distraction from focusing on the things that will actually translate to the final product (energy, emotion, style, etc.)
As for the article...
I wish the article (and people in general) deciphered a little more between correction as an effect and correction as a tool. Almost all of the examples they use are of artists who use it as an effect. There are a lot of people using it sparingly and transparently ("that take was good, if we can just fix than one note" etc.). But I guess getting in to the gray area makes for a lesser story.