Hacker News Reader

Comment by danielsju6 | Hacker News Reader

This is actually how the protocol works, it uses the certificate burnt into the CPU; Apple claims that the private keys are known by no party including themselves, probably generated by the chip fab with some some of SCEP exchange. I've been meaning to update the IMWiki page with my research, but here's the bit where you see that exchange: http://imfreedom.org/wiki/IMessage#Unknown2 ["Apple iPhone Device CA" / "Mac OS Device Identity (Production)"]

View on HN

> Apple claims that the private keys are known by no party including themselves

Source? How can I add a new device (e.g. Messages on OSX) and get all the messages between my phone and friends?

Private keys, not public; the messages are decryptable by anyone who's been added to the keybag, you've probably seen the pop up notifying you that another device (private keys burnt onto the chipset) is joining the keybag. Once you hit OK Apple pings you with encrypted communications from the other devices (them being the pipe and presumably not in the trust chain, this explains why sometimes it's a little iffy showing cross device messages.)

The protocol hasn't been completely reverse engineered but enough to know it's likely decentralized like this, just MITM it (quite a bit is documented on the wiki I linked), Apple is the CA and the piping but it appears to be rather strong and decentralized in terms of chain-of-trust. Apple has put itself, probably without coincidence, in a position where they may not even be able to execute a court order to spy on a user.

The Apple security white-papers detail their hardware level certificates, I've dealt with this a lot as a Apple MDM developer.

I've seen the message notifying me that I've added a new device to the account, yes, but what does that have anything to do with the encryption?

When someone sends me an iMessage, their client sends as many copies as I have devices?

>When someone sends me an iMessage, their client sends as many copies as I have devices?

Generally how this works is that the sender encrypts the message with a symmetric encryption algorithm (like AES) and then encrypts the randomly-generated AES key with RSA with each of the receivers' private keys. So you only send out a single message, but it can be decrypted by any of the intended recipients.

That makes a lot of sense. In that case, it seems the easiest way to wiretap an iMessage account would be to compromise one of the devices. Not surprising that the DEA doesn't have that kind of expertise/authority, though.

Nope - all they have to do is get Apple to add another surveillance key to the list of device keys associated with your Apple ID. Then all future iMessages get encrypted to that one, too.

And apple would only have to make sure, that you get no warning, that another device was added to your keybag.

They manufacture the hardware, design and produce the software, and are the intermediary between all device communication with all other devices.

I have a feeling they could manage.

I didn't mean to imply, that they could not, written it missed my ironic voice-modulation ;-)

I really had thought, that this would have happened long ago, for everybody, and so on. Would not have thought, that this was a problem for law enforcement, as it was with i.e. skype.

Whenever any iDevice downloads a new keybag with an extra key, it gives the user like 15 notifications that a new device has been added to the iMessage account. I don't think there's a way to suppress this (without releasing a new OS).

I think you are assuming too much about the coupling of the UI. Do people iMessaging me get an alert every time I buy a new iPad?

WINDMILLS DO NOT WORK THAT WAY.

No, but the person being snooped gets a notification.

Why do you assume that?

The user could detect that, though, so it's a no-go.

Not without jailbreaking or proxying the push connection...

No, by default your device notifies the shit out of you whenever your keybag is modified.

You are confusing the iMessage device activation UI with the underlying crypto operations.

Nobody iMessaging me gets any indication when I get a new iPad and add it to my account, despite all their messages to me now encrypting (on their device) to n+1 of my devices now.

If I understand this correctly, you're saying that they actually encrypt the message with a single key. Then they encrypt the messages key with the public key of all recipients and send it with the message, so each user can decrypt the message key with the private physically attached to the device and use it to decrypt the message?

Not sure I'm reading this correctly, so a quick check:

When I have A, and want to add B, A ships its private key to B so it can decrypt the messages too? (and vise versa, obviously)

Or does A get B's public key, and A then re-encrypts and re-sends messages it receives to B?

Or is it something completely different?

The messages don't always arrive in the same order between the devices, so it's definitely not passed from one to the other. I imagine the senders encrypt the message with every public key in your keybag.

Someone who actually knows should weigh in, though. GP's understanding seems a bit murky to me.

If that's the case though, where would this popup be appearing from? (I haven't seen it, not quite sure where it would be) I would guess they wouldn't be asking the sender to approve adding my multiple computers whenever I add one, since that would be a fairly crummy user experience. Seems like it should be something that I, the owner of the receiving devices, would approve or deny.

The popup (and e-mail) only notify you that a device has been added to your account. The approval is done by logging in with your Apple ID.

I'm trying to find more info about the certificate burnt into the CPU. Thus far, all I can find is the UID, which is an AES key burnt into the CPU and protected from cryptanalysis. However, I can't find anything about an asymmetric crypto key burnt into the CPU. Do you have any more info on that?

http://images.apple.com/ipad/business/docs/iOS_Security_May1...

You can find more info here. Encryption and Data Protection section should be a good place to start.

I've already read that document. There's nothing about burned-in asymmetric crypto keys, only the burned-in symmetric AES key (the UID).

That begs the question: how does my private key securely get shared between my mac, iPhone, and iPad so that all of my devices can securely read messages delivered to me.

The protocol appears to toss around the keybag, which causes the devices to re-sign payloads and toss them back up to Apple to pipe around—I've responded in more detail on a sibling.

Reply on news.ycombinator.com