pull down to refresh
My take assumes the mixing works.
But the underlying motivation was the shit storm nvk and airgappers say how coldcards and others protect exfil.
We'll son the exact thing you bitch and moaned about could saved your butt.
I really like Bitbox perspective on airgapping.
https://blog.bitbox.swiss/en/does-airgap-make-bitcoin-hardware-wallets-more-secure/
I could make the counterargument that the usb cable providing host entropy adds complexity to the RNG code, more places for bugs. And now we have to worry the software on host is sending good entropy, and that same untrusted host is not sending /malicious entropy/.
Complexity is the worst enemy of security.
While I agree that more entropy is better, in this case we add an attack surface and code complexity for something that can be handled entirely on device. Have the device use sufficient entropy, and use cryptographically secure algorithms.
This device sw is open, and could have been discovered any time. Finally someone found it. Because the 1st batch went to a single wallet I hope the situation ends up being whitehat rescue, and people get refunded.
Weak key generation in Bitcoin wallets is an evergreen problem. Analyze your favorite wallet today!
I see your point, but just because a connected wallet can mix entropy sources doesn't mean it will be implemented correctly.
Airgap or USB, this seems like an issue of bugs escaping notice and a new actor coming on the scene that notices a lot more.
I wont be surprised if there is another one of these within the year: some major hardware (or software) wallet provider that has some very nitty bug in it that breaks everything.