There are definitely use cases where software engineering can bring about harm if done poorly/maliciously, but they're outnumbered by ones where it really won't/can't. How many people are working on those things you list above compared to say, websites or CMS plugins or video games or marketing banner or emails or what not?

That's another huge difference between software engineering/web development/coding and something like medicine; there are few parts of the latter where screwing up won't have major consequences, and many parts of the former which won't.

In that sense, I guess you could say tech is less like a 'field' or 'industry' and more like part of life that a bunch of different fields or industries depend on. It's like how 'writing English' covers lawyers writing documents for use in court, teachers teaching students in schools and colleges, authors writing novels and businesses writing copy for their websites/brochures.

The former two cases would have some sort of licensing because they have actual consequences if someone doesn't know what they're doing, whereas the latter clearly wouldn't.

It's cool. They will have one proper "engineer" to verify and sign off on everyone else's cheap and/or shoddy labor.

Which recent bootcamp grads are writing spacecraft software and cars?

That isn't the point. The point is that it's not the case that software failures are inconsequential. I'm sure there are some bootcamp grads who would like to work at SpaceX; I happen to have a personal contact who I wouldn't trust to code his way out of a paper bag who just got a job at Equifax.