There are different definitions of "low level language", but I think a charitable interpretation of the one in this article is that they view processors as providing a virtual machine, and count "levels" from what's more efficient for hardware (that is, the bits below that virtual machine). Though maybe a somewhat strange/confusing title was picked to attract attention.
OK, following that, in the case of the GPU, C++ code is translated into SPX "assembler" but SPX is very much a macro system and the programmer doesn't get access to the true low level.
And I don't think we'll escape the situation that there will low-level emulation code that programmers can't and should not access. It's good to know but that doesn't change the levels programmers normally work with.
