Apple acknowledges tracking iMessage metadata and sharing it with law enforcement
Despite strong encryption -- and claims that it "doesn't scan your communications" or "store data related to customers' location" -- Apple is saving some metadata from iMessage and other apps and sharing it with law enforcement agencies, according to a new report.

In a document about Apple's iMessage system obtained by The Intercept, the Florida Department of Law Enforcement's Electronic Surveillance Support Team noted that when users enter a phone number into iMessage, metadata is periodically uploaded to Apple servers to check whether a text should be routed through iMessage or standard SMS. This material includes not just phone numbers but the date and time of the lookup, and the querying user's IP address.
While the data doesn't include message contents, or even reveal when conversations happened, it could potentially be used to identify who a person is associating with, and/or trace an IP address back to a real-world location.
Responding to The Intercept, Apple acknowledged the data collection, saying that it retains logs for 30 days and hands them over when served with a valid legal request. Because these orders can sometimes be extended in 30-day blocks, though, it's possible that some people are being tracked for longer durations.
"When law enforcement presents us with a valid subpoena or court order, we provide the requested information if it is in our possession," Apple said in an official statement. "Because iMessage is encrypted end-to-end, we do not have access to the contents of those communications. In some cases, we are able to provide data from server logs that are generated from customers accessing certain apps on their devices. We work closely with law enforcement to help them understand what we can provide and make clear these query logs don't contain the contents of conversations or prove that any communication actually took place."
Apart from Messages, the company didn't specify which apps are uploading metadata.
Though Apple is often considered more invested in privacy than other high-tech corporations like Google, it has regularly complied with data searches by U.S. law enforcement and spy agencies. In 2013 it was implicated in the National Security Agency's PRISM program, found to be gathering customer data from a number of American tech companies. Apple denied providing "direct access" to its servers, or even hearing about the program.

In a document about Apple's iMessage system obtained by The Intercept, the Florida Department of Law Enforcement's Electronic Surveillance Support Team noted that when users enter a phone number into iMessage, metadata is periodically uploaded to Apple servers to check whether a text should be routed through iMessage or standard SMS. This material includes not just phone numbers but the date and time of the lookup, and the querying user's IP address.
While the data doesn't include message contents, or even reveal when conversations happened, it could potentially be used to identify who a person is associating with, and/or trace an IP address back to a real-world location.
Responding to The Intercept, Apple acknowledged the data collection, saying that it retains logs for 30 days and hands them over when served with a valid legal request. Because these orders can sometimes be extended in 30-day blocks, though, it's possible that some people are being tracked for longer durations.
"When law enforcement presents us with a valid subpoena or court order, we provide the requested information if it is in our possession," Apple said in an official statement. "Because iMessage is encrypted end-to-end, we do not have access to the contents of those communications. In some cases, we are able to provide data from server logs that are generated from customers accessing certain apps on their devices. We work closely with law enforcement to help them understand what we can provide and make clear these query logs don't contain the contents of conversations or prove that any communication actually took place."
Apart from Messages, the company didn't specify which apps are uploading metadata.
Though Apple is often considered more invested in privacy than other high-tech corporations like Google, it has regularly complied with data searches by U.S. law enforcement and spy agencies. In 2013 it was implicated in the National Security Agency's PRISM program, found to be gathering customer data from a number of American tech companies. Apple denied providing "direct access" to its servers, or even hearing about the program.
Comments
And before someone says it...this is COMPLETELY and UTTERLY different from the iPhone 5c case with the FBI. Not even close!
As as long as the messages themselves aren't logged.
So, would these metadata hold up in court? All a defense lawyer would need to do is provide this excerpt, which is straight from Apple.
If anyone follows the Adnan Syed case, you know that an important statement, similar to the one above, from a recently discovered (i.e., discovered after the trial) cover sheet from AT&T has thrown a huge monkey wrench into the state's (Maryland) case, which hinges on incoming phone calls supposedly received by the defendant in specific locations. The cover sheet basically says, "Incoming calls are not reliable for pinpointing location." Ooops.
It doesn't for example prove receipt of a message, only that a particular identity was looked up in the "To" field. That would typically be more useful to police as intelligence (i.e. information that influences decisions in an investigation, but does not directly impact the argument to prove guilt) , than as evidence.
The original article is at an almost absurd level of paranoia - similar to "Banks track customer spending patterns" or "Grocery store knows what's in your shopping basket" or "Post Office knows when you have mail".
The central directory of identities , that holds public keys, and deals with the issues around multiple devices per user, has been a key feature of iMessage & why it's been easy to use since launch.