I don't enjoy arguing, but Andrew Tanenbaum had a choice, when re-licensing Minix, to go with GPL or some other copyleft license.
GPL won't stop bad people from doing bad things, but the thing with that is that you have to show your modified source code in unobfuscated form and with full access (to all paying customers, at least). It keeps companies honest. Intel can't pull this off with the Linux kernel for Intel ME, because they'd have to publish its full source code. Everyone who understood the code would realize that Intel's doing some shady shit.
In contrast, Minix, by being permissively-licensed, doesn't have this "restriction". Intel modified Minix for Intel ME because they don't have to publish any source code of their own. At least, that's what I'd do if I had something to hide. Intel happily contributes to Linux regularly, why not open-source Intel ME, right? Good security can exist in open-source; look at all the perfectly-good open-source cryptography libraries out there. That they're not is a deafening silence. Apple uses the BSD system (it's not just a kernel), not Linux, because they don't have to do that whole "publish source code" thing, and look at how locked-down their shit is.
I'm on the fence between permissive and copyleft for this reason. Every contributor to Minix has blood on their hands because of Intel, depending on how far the chain of blood applies. They all played a hand in it. Would you want to be Einstein, having discovered that your life's work played a part in the nuking of Japan in WW2?
I dunno, it's just my personal morality. I don't want to be responsible for bad things, no matter how dilute that responsibility is. I suppose you could say that I'm indirectly responsible for a lot of bad things, by that logic. It's not rational, I will admit.