Comment by marcprux

It is possible to set up end to end encryption where two different keys unlock your data. Your key, and a government key. I assume google does this.

1. encrypt data with special key 2. encrypt special key with users key, and 3. encrypt special key with government key

Anyone with the special key can read the data.the user key or the government key can be used to get special key.

This two step process can be done for good or bad purposes. A user can have their key on their device, and a second backup key could be in a usb stick locked in a safe, so if you loose your phone you can get your data back using the second key.

Is the source code for every binary blob present on an Android device available for inspection, and is the code running on every Android device verifiable as having been built from that source?

> or through convictions

If they wanted to use this evidence for a normal criminal case, they would just do parallel construction.

It's worth noting that what the security services don't have access to is as secret as what they do have access to. According to the late Ross Anderson, for many years the police were unable to trace calls (or was it internet access?) on one of the major UK mobile networks, because it had been designed without that and in such a way that it was hard to retrofit. This was considered highly confidential, lest all the drug dealers etc switch to that network.

Would it be possible that they feel that the revelation of this backdoor would be too big of a loss so that any of these theoretical cases of the past 7 years have used parallel construction to avoid revealing the encrypted data was viewed?

> Wouldn't there have been be some evidence of that in the past 7 years, either through security research, or through convictions that hinged on information that was gotten from a supposedly E2E-protected backup?

I wouldn't count on it. The main way we'd know about it would be a whistleblower at Google, and whistleblowers are extremely rare. Evidence and court records that might expose a secret backdoor or that the government was getting data from Google that was supposed to be private could easily be kept hidden from the public by sealing it all away for "national security reasons" or by obscuring it though parallel construction.

I don't know the particulars, but in general, silence around a massive tech company on warrants does not mean "they said no and the feds decided to leave them alone"

A trivial method for circumventing code review is to simply push a targeted update of the firmware to devices subject to a government search order.

There are no practical end-user protections against this vector.

PS: I strongly suspect that at least a few public package distribution services are run by security agencies to enable this kind of attack. They can distribute clean packages 99.999% of the time, except for a handful of targeted servers in countries being spied upon. A good example is Chocolatey, which popped up out of nowhere, had no visible source of funding, no mention of their ownership structure anywhere, and was incorporated along with hundreds of other companies in a small building in the middle of nowhere. It just screams of being a CIA front, but obviously that's hard to prove.

[dead]

[dead]