I agree, and I want to add that 'better' doesn't necessarily mean 'creates more robust, elegant, resilient software'. Better means from a business perspective. If we (I'm one of the people you're discussing) end up cheaper or more fungible, for example, we still might be worth hiring from a business perspective even if the code we create is shit.

I've also seen an assumption that you've made here that I think is worth drawing attention to and questioning: that the tech-savvy non-developers are starting from zero or near zero when it comes to programming and software development. Right now, that's probably mostly true, but I'm not sure that will continue to be the case. I'm not a developer (depending on how fuzzy we want the boundaries around the idea to be, anyway). I do understand the building blocks of programming languages (e.g. I can answer all the questions fragmede posed in a sister comment), the trade-offs between rolling your own and using existing libraries, the need to evaluate tools, frameworks, and languages to determine which is best for your use case, why version control matters, why access rights matter, why backups and a test environment are necessary, why it matters to write code another human can read, etc.

Do I understand as much as an active working developer? Absolutely not and I'd never claim to, but I'm far from starting at zero.

The reason for this is that I was raised by programmers. There are far, far more programmers and general tech nerds now than there were in 1988 (when I was born). Which means that in 10-20 years, there are going to be a lot more children, grandchildren, nieces, nephews, and so on of developers, and a lot of them are not going to be starting at zero. For pretty much of all computing history, there's been a substantial opportunity cost to developing a deep understanding of coding and software development: either a person has to be so into the domain that they devote a lot of their waking hours to it (usually in adolescence or young adulthood, when that trade off closes the most doors and makes developing certain other time intensive skills difficult), or they have to obtain a CS degree, which means not getting a different kind of degree and often incurring significant front-loaded financial costs. The opportunity cost for people born into programming or tech families is much lower. You can start younger and spread out the hours needed to learn across a greater amount of years, you can acquire knowledge in less time-intensive ways and while practicing other skills (e.g. my cousins also have 'software brain' and we could all hang out and develop those skills while also developing in person social skills), and you have a built in network of experienced people who want to help you + that can give you extremely individualized, personalized attention.

If what you suggest comes to pass, I think that one of the greatest threats to SDE as a career is going to be your own children and grandchildren.