Amazon and other tech companies can design systems where the keys are held entirely by the end-users. How far is the US willing and able to go to get at that data?
How? If I send an e-mail via Amazon's service, they must at minimum have the standard e-mail headers in order to process it. In which case, it's no different than PGP-encrypted e-mail, available with almost any e-mail provider. This also provides Amazon (and whoever subpoenas Amazon) with significant meta-data on my communications, even if they don't have my actual text.
In order to make e-mail into a system where two people can communicate in a secure fashion, where no data is stored on or passes through a remote system unencrypted, you would have to re-implement e-mail. And of course there goes interoperability.
