Raspberry Pi Pico cracks BitLocker in under a minute
theregister.com
theregister.com
Anyway, not to detract from the nice work of the author or to tout my own horn but I can hack a lot of encryptions in seconds with a simple keylogger. For the sake of this exercise I'll consider the key exchange (user typing password) is an integral part of any encryption scheme :).
More seriously, I think fTPM or TPM+PIN+USB key would be good ways to avoid this scenario.
Except you can use good old password protected FDE on any major desktop operating system other than Windows Home Edition.
> I can hack a lot of encryptions in seconds with a simple keylogger.
A thief can't use your keylogger to decrypt a stolen laptop that's properly encrypted. A rogue recycling shop can't do that either. And kids won't be able to use, uh, Raspberry Pis to decrypt random Surface laptops.
So yes, it's kind of a big deal. People shouldn't have to worry about the TPM details of their devices to benefit from encryption in ways that protect against the most common threats.
But one could not do that and be just as vulnerable. Same as you can use a PIN on BitLocker and be safe in this case. Which answers my question that this is not something BitLocker specific, just a bit of name dropping to garner more attention. Which is fine from the original author, not so fine from the journalists that picked it up.
> So yes, it's kind of a big deal.
You're not wrong but also this is not new. On close read this is indeed the same attack detailed as far back as 2019, says the internet. Presumably a few dollars cheaper now.
Use an additional factor if you care about that data staying encrypted, it only takes a few seconds and could save you quite the headache.
Right, but BitLocker is vulnerable by default whereas LUKS is not, because BitLocker relies on a TPM without a PIN by default whereas LUKS relies on a password by default. Yeah, technically this attack ain't BitLocker-specific, but I don't know of any other FDE implementation that defaults to using a TPM without a PIN/password.
[0]: https://news.ycombinator.com/item?id=39243305 [1]: https://news.ycombinator.com/item?id=39284711
Mission critical embedded devices are not designed by Microsoft, and would normally have tpm parameter encryption...
https://security.stackexchange.com/questions/253776/why-does...
Turns out it's all security theater.
There's a mode of Bitlocker where it boots into a basic boot environment, asks the TPM for the key, the TPM validates the environment, and then gives the decryption key. For some discrete TPMs, this last step of the TPM giving the key to the boot environment is done in the clear and can be sniffed.
Or, encryption passphrase on boot.
But in that case, you just need two accesses: Add h/w keylogger, read h/w keylogger.
You can at least make it somewhat more difficult by using ubsguard to prevent the most obvious keylogger ingress points.
However, as tech people we need to stop downplaying our accomplishments. “43 seconds” (plus the lifetime of learning that allowed them to figure out how to do all these things: decoding the wire signals, writing a custom firmware, knowing how to probe the motherboard for the correct signals, etc.)