Hardware wallets are NOT part of "Bitcoin security chain". They are tools created by centralized companies and sold to anyone who wants the utility they offer. Anything that happens to them is happening to the product of that centralized company NOT to Bitcoin.
My understanding is that security needs to be evaluated holistically, encompassing how people interact with the system. I would suggest that key management is a major element of the “Bitcoin security chain”.
Let me put it this way: If the majority (or a significant proportion) of BTC is controlled via keys generated and stored in HW wallets, then I think my point still stands. If HW wallets are compromised then it would affect the project as a whole.
Again those chips are not related to Bitcoin.
They are related if they are predominant in key management.
If you want to describe an attack you first have to think about costs and benefits. What would the "state actor" gain by carrying out this type of attack? What would be the costs and most importantly why not use any other alternative method that doesn't cost as much and are easier to pull off?
The attack I described above seems to me to be relatively low cost and high benefit. If state actors have the ability to generate non-random keys, or to leak keys via subliminal channels the they effectively have control over the entire Bitcoin monetary system, worth about $1.2T.
To be clear, I’m not saying they would just steal everyone’s BTC in one go. But control over the majority of keys would allow them to target individuals, for example, in a way that could be covered-up as some kind of op-sec failure. In any case it doesn’t matter: the ethos should be “don’t trust, verify”. The “system” as a whole, must be provably secure.