In that they know they're making a choice with consequences and pain. But they see the strategic value of that choice (never running a support org) as outweighing the tactical pain.

Additionally, I'd guess the internal opinion is "If users need support, then something is broken and we should just fix it."

Where I think this falls down, which Microsoft learned 30 years ago, is that:

(a) Unless you're asking, you're not going to surface visibility of broken things. Users will just find a way to deal, while being pissed off. And then some product team finds out there's been a major issue for the last 5 years.

(b) Unless you're actively soliciting feedback from your customers and users, you're at high risk of building the wrong thing. We see this with enterprise GCP frequently. "Oh, you need Y feature? We never thought about that, because {insert internal way that Google does things, but no one else does}."