NOTE: This post does not address the legality/constitutionality of the warrant or the intent of the party who requested it. The sole point of this post is to determine exactly what the warrant can realistically achieve, based on engineering/technical limitations and assuming Google’s total compliance.
Google was recently served a warrant by law enforcement, requesting that they assist in unlocking a suspect’s Android phone that was seized on January 17, 2012. The phone was locked via a pattern password. Here’s what the warrant requested, with my comments on some of them to the best of my knowledge:
- The subscriber’s name, address, Social Security number, account login and password
Assuming there’s some way to access the device ID of a locked phone, I suppose Google could get to the account login at the very minimum. AFAIK Google passwords are stored in encrypted form, which means even Google can’t reach them. Google doesn’t require a real name or address – or at least they can be easily spoofed – for account creation. Also no current Google service uses SSNs.
Google might be able to force a password reset, but I’m not sure if that works remotely for a phone as the device would have to receive the updated password from Google’s servers. IMO unless remote password reset works, then there’s no way to unlock the phone without the owner’s compliance.
- “All e-mail and personal contact list information on file for cellular telephone”
With the account login, email can be accessed. Personal contact list can be accessed by Google remotely (without access to the phone) if the suspect synced his contact data with Google Contacts. Otherwise, it’s on the phone only and requires unlocking to be read.
- The times and duration of every webpage visited
I think remote access for this works only if history syncing is enabled on the phone or in the browser app. Otherwise it requires unlocking the phone too. Carriers do store IP destination information, but this handy data retention chart shows that only Sprint and Verizon would be able to recover anything from before the phone was seized:
UPDATE: According to the application, the phone model in question is a Samsung SGH-T69, a T-Mobile device, so it looks as though recovering web history from the carrier is a dead story
- All text messages sent and received from the phone, including photo and video messages
AFAIK Google doesn’t store texts (though some of its apps may read them). I’ve seen some comments suggesting that this could be obtained from the suspect’s carrier, but the abvove data retention chart shows only 1 carrier – Verizon – stores SMS content at all, and they do it for a maximum of 5 days only. All carriers store SMS details for at least a year, however, so that might be a productive avenue.
- Any e-mail addresses or instant messenger accounts used on the phone
Assuming the device ID is accessible, the email addresses can be pulled. IM account details require unlocking, especially if the suspect used 3rd party IM apps.
This is a pretty interesting case both to law enforcement and private citizens alike, it’ll be enligthening to see whether Google can unlock the phone and what data they can come up with if they can or can’t.
UPDATE: Law enforcement may already have part of what they need. The application says they already have the device’s IMEI, which from my reading appears to be used by Google to tie user accounts to devices. IMEIs are device specific and (usually) unique. It doesn’t give them the data that’s stored on the phone, but it does give them everything that’s on Google’s servers under the matching account name. Of course, if the phone is registered to a Google account other than the suspect’s then the Google server data might not be useful.