On the contrary, I think your example proves my point quite well. I understand very little about PDP-11s and only slightly more about GPT-3's inner workings, yet I have no trouble reasoning about whether or not a PDP-11 is suitable for running GPT-3 or something even more difficult to formally reason about, say Microsoft Windows. I have a mental model of computer performance and compute requirements that simplifies the question from a difficulty of "Oh, it's Turing complete, halting problem, let's throw our arms in the air like this is an infomercial!" through "You need to understand literally everything about PDP-11s and Windows" all the way to "50 years of exponential growth is a hella large factor to try squeezing down anything by." It's a trivial question hiding in the skin of an intractable question, and it perfectly exemplifies why it's silly to believe that human cognition will forever remain intractable.

In order for a problem to forever remain in "let's throw our arms in the air like it's an infomercial" territory, it must not merely be difficult in its most pedantically defined complete form, it also must stymie the search for useful relaxations and workarounds. Nobody fears running a program on account of being unable to prove that it will halt: they just kill the program if it locks up, or (equivalently) set a timeout. Personally, I'd just avoid throwing my arms in the air like an infomercial altogether.

EDIT: substituted GPT3 -> Windows because arguments about GPT-3 and/or a set of incarnations being Turing Complete would be irrelevant to the main point.