It had already happened: http://www.wired.com/threatlevel/2010/09/google-spy/

Google acknowledged Wednesday that two employees have been terminated after being caught in separate incidents allegedly spying on user e-mails and chats.

David Barksdale, 27, was fired in July after he reportedly accessed the communications of at least four minors with Google accounts, spying on Google Voice call logs, chat transcripts and contact lists, according to Gawker, which broke the story Wednesday.

Google has acknowledged that it fired Barksdale for violating company privacy policy, and acknowledged that it was the second such incident of its kind at the company.

It doesn't matter, the information is there. He mentions building a profile. While the machines might have built it, there's nothing stopping a real person (forced or otherwise) from browsing those profiles.

First of all, there are lots of things stopping them (systems permissions, checks, organizational rules, etc). All might be circumventable -- but to claim they don't exist is ridiculous.

Secondly, intent matters. Killing a person can be a pure accident, can be manslaughter, can be first degree premeditated murder. You might attempt to dumb it down to "It doesn't matter, the person is dead" -- but that isn't how the US legal system, or the majority of people think.

If the intent of a system is to allow the profiles to be read by people (or shipped to government) that is very different than if the intent it to be exclusively used by software.

"The information is there" on all e-mail providers. There is nothing (short of end-to-end encryption) stopping a real person at any e-mail provider from looking at your email.

The interesting distinction (for me, and I have been a happy GMail "customer" for close to a decade soon) isn't the profile, it's the human. I have strong faith that Google can be successful in keeping humans from reading my e-mail (or a computer-generated digest of it, which is what the profile amounts to), and so I don't worry about the existence of the profile.

Not always. Several new security-oriented email providers boast diskless email servers and end-to-end encryption.

Yes, end-to-end encryption would fix all this, but it would also fix it in OP's satirical mail service, as well as in GMail - so that's really a tangential point.

Trusting that those providers don't start siphoning off a copy of your mail (or are indeed diskless and not out of malice or incompetence actually just using regular disks. Also, being diskless is worthless if they are still keeping your mail around in memory anyway) is no different than trusting Google to keep humans away from my email.

Why would you think they would hire a human to read mails? As I understand you submit your message online, they analyze your message the same way emails are analyzed, then they print the letter along with some ads and puts it in letter. Hiring people to read the letters would be extremely inefficient.

> Machine "reading" the email != humans reading the same thing

Try saying that in any of the gajillion PRISM threads.

Machine "reading" the email to glean useful advertising data != machine "reading" the email to find any excuse to label you a terrorist.

> machine "reading" the email to find any excuse to label you a terrorist

Is there any evidence at all that this is happening? So far They don't need to do that, they use corrupt informants (Guantanomo) or stupid videos on YouTube (US teen "rappers").

Yes, there is. [1]

[1] http://www.washingtonpost.com/blogs/the-switch/wp/2013/08/05...

>A machine can't laugh at your embarrassing confession. Nor can a machine talk about your letter to its friends.

To its friends at the NSA? I beg to differ.

That's completely not the intent of the quoted text and you know it. There's certain implications with that phrase. Blindly sending data != talk about your letter to its friends.