Yes it is , and I have read those so-called " fixes " elsewhere , and some are very risky ..... " Sledgehammer to crack a nut " .Minterator wrote: .... It's in the chipset, not the CPU ......
And as you correctly say , the problem is mobo based , and we are going to be stuck with that situation for a long time I fear .
This is the best starting point that I can think of -
To stand back , and get a full picture of how bad the problem really is , with hardware that we have implicitly trusted until now ,
( foolishly , as it turns out )
"Towards (reasonably) trustworthy x86 laptops " by Joanna Rutkowska ( Invisiblethingslab /Qubes )
She addresses these crucial questions -
"Can we build trustworthy client systems on x86 hardware?
What are the main challenges?
What can we do about them, realistically?"
The fundamental question that comes out of all this :-
Is it actually possible to have a truly stateless computer ?
And the answer right now , at least as far as Intel hardware is concerned , is NO , it is not .
The scenario with AMD might be a different , I don't know , but for now let's assume that it is equally dismal ,
until proved otherwise .
Update : Not good news about AMD ; they have something called " Platform Security Processor " ( aka " Secure Processor " ) .
This is from 2014 , but it's interesting to see that the diagrams ( graphics ) look very similar to those for IME .
http://www.electronicdesign.com/micropr ... power-apus
And more recently
https://hothardware.com/news/amd-confir ... essor-code