As for the attack, imagine a browser that encrypts local storage with a system key. If I understand correctly, by storing different patterns of bits, Hertzbleed might be able to extract the system key from the timings to save data.
This might sound very theoretical, but modern OS'es (and password managers) have lots of keys like that. There's a good chance one or more of them are reachable from Javascript. And that's just what popped in my mind in two minutes, I'm sure red teams will have better ideas.
The scary part is that this is another attack in the same ugly class as Meltdown and Spectre, where the antidote is nearly as damaging as the poison.
Don't you need precise clocks for this in JS? The ones that were disabled in browsers after Meltdown/Spectre.
Isn't that practically impossible given the amount of random software running on a regular computer nowadays? This is not based on hdd write speeds but processor timings which are affected by every single cat video you are watching.
For some single threaded code we force other cores to idle so that one thread can get maximum cache and maximum frequency. A similar approach could be used to minimise the power analysis signal.
short secrets which aren't mitigated by MFA: session cookies, TLS client certificates, secret keys of e2ee IM apps (eg. Element), Zerobin URLs, ... Maybe even TLS session keys?
I can see that if other persistent access credentials were sitting unencrypted in memory that could be an issue if the attacker can use those to hop into other services the user has access to, but that's manageable via care with device usage, proper MFA etc right? Or I can see how true external IP address might be a dangerous short secret for someone trying to hide (vs merely secure) via VPN. But I think in those cases externalizing the VPN to the network level is a good idea anyway since being 100% sure of zero leaks in a modern OS/application stack is challenging, and then the user device doesn't have to know anything about the real IP at all. JS also already allows a huge amount of fingerprinting, so if someone is allowing JS to run and is deadly worried about privacy they must be thinking about mitigations there already.
Again, not at all denying that incredibly tricky stuff can be done using little leaks like this to escalate in surprising ways. But for a dedicated web browsing device with only outgoing and session access to WAN, likely though a VPN but not necessarily on-device, what new threat scenario is this such that completely disabling dynamic frequency is the only response? Although I suppose for a dedicated web browsing device using tor browser or the like disabling that might not actually be a big deal anyway.