If it means doing more deals like when Obsidian was allowed to make New Vegas, might finally mean that Bethesda franchises get some decent entries again.
I think you've hit the nail on the head. They clearly skimped on the content for all those planets and handwaved it away as perfectly normal and expected, when it clearly wasn't good. Why? To add breadth without adding time to the development. To "increase their output" without adding more input.
It clearly didn't work there, but it could have worked with some decent mechanics and a little more thought into the content. It worked for No Man's Sky.
Luckily for Bethesda, AI has suddenly gotten a lot better, and they'll be able to use it to generate a ton of content that feels better than standard old procedurally generated content. That is, of course, if they can manage to work it into their tooling for their ancient engine.
It's really hard to fill a space game with content. It's not that surprising that a lot of the world's are empty. This is an issue for all space games, not just Bethesda.
Which is why smarter devs either keep all the action in space, or limit it to specific places in specific planets. Besides, do we really need to land on literal hellscape planets like Mercury or Venus?
By devs I meant developer studios in general, not the actual coders.
Emil Pagliarulo and Todd Howard are pretty much the two "they say it, you do it" voices in Bethesda and, as far it's been shown, Microsoft was very hands off with how BGS handled Starfield.
In this specific case, it really looks like it was a case of terrible design decision from high up, either Todd or Emil, to "let the player land on every solid rock" and have half of them have human buildings
As a comparison, Elite Dangerous, which is not AAA, but as close to mainstream as a space game gets, is a game about space activities, including exploration, and it took ~6 years to release a DLC that added planetary landing, and that was super limited, too.