The answer to your question is in the next sentence, "While the original codebase is MIT, any changes here are GPLv3. Supporting closed devices with this code is not a goal."

That's the entire point of a permissive license. You grant other parties the right to do as they wish, given a very lax set of requirements (usually just attribution). Would you feel offended if someone only distributed a binary build with their changes, or even charge for that? That's literally what the permissive license is about.

What may not be immediately obvious, is that it still favors the downstream developer to contribute their changes back; the network effect scales exponentially with the number of participants. Making the changes copyleft complicates the matter, as their copyright status may become less clear as further downstream contributors chip in.

There's always some trade-off, but this is exactly what the MIT license permits. If it bothers you, you'd probably find yourself preferring copyleft anyway.

Protection against corporate interests taking over the project in the future.

Usually ideology. GNU/FSF people love switching software from permissive to copyleft.

The dude that worked on it prefers GPL over MIT.

Not less permissive, just targeting the end user instead of the developer rights.

I think they want assistance with the project, not for corps to take it and build it without contributing back. It's why Linux has grown more feature/driver rich while others grow at a slower pace (no necessarily a bad thing).

There’s no argument for the change in the general case. There’s rationalizations, sure.

But no euphemisms like “public”, “free software”, “copyleft”, make the change anything less than taking away a freedom and calling it “free as in speech”.