Your statement appears to be false. I imagine you’re responding to this article — if so, perhaps you should take another look at it.

The next iteration of Python, 3.8, has nothing like this. It’s coming out towards the end of this year.

Python 3.9, due in mid-2020, will support sub-interpreters. In short, this means multiple copies of Python, each with its own GIL, which can use techniques like shared memory to communicate.

This does avoid a lot of the issues associated with the GIL, but it certainly doesn’t remove the GIL — in fact, it multiplies it. Conceivably, it could be a step to later removal of the GIL, but that isn’t even on the roadmap.

The problem I discuss in the article you are responding to is still there — you need a way to make sure that fundamental Python data structures like lists and dictionaries stay consistent even if they are being used from different threads. Some form of synchronization is needed for this and today that’s the GIL.

I could be completely misinterpreting your comment, but given you present no argument or references, I’m forced to guess what you mean.