18

Edward Snowden's official Twitter account posted the following tweet:

enter image description here

The first 4 points are not in contention, but the 5th point states that the FBI does not require Apple's assistance to unlock the phone, implying that the FBI is using this case not to gain access to the phone, but instead to set precedent, as Apple claims, and as FBI denies.

Alternative means for gaining access to this device -- and others -- exist that do not require the manufacturer's assistance.

Is there any publicly known information corroborating this claim?

  • 4
    No respectable government would have a problem cracking a device it had physical access to. Do you think the entire FBI can be defeated by one iPhone? – D J Sims Feb 29 '16 at 1:04
  • May be this could help somebody form an answer-crypto.stackexchange.com/questions/32886/…, and a user named spdustin comments with references that access to a computer synchronized with iCloud could have helped in extracting the backup without Apple's involvement-news.ycombinator.com/item?id=11199093! – pericles316 Mar 1 '16 at 14:27
  • The second method is decapping which is mentioned by Snowden himself here-ibtimes.co.uk/…. – pericles316 Mar 1 '16 at 14:38
11

There was a discussion at security stackexchange about this issue.

The top rated answer says:

Yes, it is possible. However, that runs the risk of destroying the device without getting the data off first, which is undesirable. It also does not achieve the political goals of forcing Apple to assist in decrypting the device, paving the way with precedent for the flurry of future requests of this sort to come, some of which are certain to have less favorable facts and thus are not as suitable as test cases.

If Apple helps, the FBI gets the data directly. Otherwise they have to invest more resources and risk damaging the phone in a way that make the data unable to be recovered.

Update (2016-09-21 11:21):
We now know that the FBI accessed the iPhone in question without help from Apple and there might be even more straightforward and cheaper ways than the one the FBI used. The register wrote a story titled FBI overpaid $999,900 to crack San Bernardino iPhone 5c password:

University of Cambridge senior research associate Sergei Skorobogatov has laid waste to United States Federal Bureau of Intelligence (FBI) assertions about iPhone security by demonstrating password bypassing using a $100 NAND mirroring rig.

  • 4
    Ummm, sorry, that's not enough evidence. -1. – George Chalhoub Feb 24 '16 at 14:08
  • 6
    @GeorgeChalhoub : "How" is never a question that matters much on this website because this website is not about original research. Consensus of the security stackexchange community is stronger evidence than Snowden's assertions and thus works under the prevailing standards of evidence. – Christian Feb 24 '16 at 14:28
  • 3
    Again, all I see is what I call apple encryption/decryption conspiracy theories so far. Everyone has his opinion about encryption or about Apple's iPhones. No, an up-voted answer on the sister site is not enough evidence. Lots aren't educated about Secure Enclave. So, no conspiracies theories and opinions aren't welcome here. – George Chalhoub Feb 24 '16 at 17:07
  • 3
    Please, Rob, read this document: apple.com/business/docs/iOS_Security_Guide.pdf which contradicts directly what you are saying. You cannot decrypt the iPhone data outside the phone unless you can crack SHA-256. Which nobody can. – gnasher729 Feb 24 '16 at 20:49
  • 2
    @Rob "so its a matter of time" - Yes. But for some encryption algorithms in common use today, it's a matter of time on the order of the age of the universe. – Kevin Krumwiede Feb 25 '16 at 9:15
7

The answer is yes, there are alternative means for gaining access to the Apple Iphone 5c device running on IOS9 that exist which might not require the manufacturer's assistance which is agreed by security expert sites such as this.

Some of the alternative methods for gaining access to the device are

Method #1: Accessing a computer synchronized with iCloud which could have provided an iCloud-specific token that might have been used in extracting the backup with reference to user named spdustin.

Precisely - from where I'm sitting, the disregard for standard forensic examination procedures shown by the reset of the iCloud password proves that their actual desire to obtain the data is not in proportion to their assertion that Apple gives in to their demand. Additionally, had they brought the device to a known access point and plugged it in to a charger, they could have availed themselves of another handy thing: Access to a computer synchronized with iCloud would've yielded an iCloud-specific token that could be used to download and extract the backup (without Apple's involvement), bypassing even TFA. Source: Ycombinator news

However, the Icloud password for the required device has been reset by FBI rendering this method redundant.

Method #2: Accessing data with the use of acid and lasers mentioned by Edward Snowden. However, this method has been only tried in a microcontroller named Infineon SLE 66PE carrying the TPM or Trusted Platform Module designation of security utilized in Xbox 360 and never in an Iphone.

Chip decapping is a mechanism where the main processor chip of the phone is physically attacked to probe its contents. First, acid is used to remove the chip's encapsulation. After that, a laser drills down into the chip in an attempt to expose the portion of the memory that contains the iPhone's unique ID (UDID) data. Tiny probes are then placed on the spot where the data is to read out the UDID bit by bit, as well as the algorithm used to untangle it. Once the targeted data has been extracted, the FBI can put it on a super computer and gear up to recover the missing pass code by simply trying all possible combinations until one unlocks the iPhone data. Since the process is being done outside the iOS, there is no 10-try limit or self-destruct mechanism that can wipe the data. Source: IB times

Also, the above method is very risky as even a small error in the decapping or attack process could destroy the AES-256 chip and the phone memory's access could be lost forever. Source: idownloadblog

Method #3: Copying the A7 chip multiple times and trying to enter combinations to brute force the PIN with reference to California Republican Congressman Darrell Issa's questioning the Federal Bureau of Investigation director James Comey .

There’s only a memory and that memory—that non-volatile memory sits here—and there’s a chip, and the chip does have an encryption code that was burned into it. And you can make 10,000 copies of this chip, this non-volatile memory hard drive, then you can perform the attacks as you want on it. Now you asked specifically Apple to defeat the finger code so you can attack it automatically so you don’t have to punch in codes. You’ve asked them to eliminate the 10 [attempts] and destroy. But you haven’t ask as far as I know asked them, “OK if we make a thousand copies or two thousand copies of this and we put it with the chip and we run five tries, 00 through 04,” and throw that image away and put another one in and do that 2,000 times, won’t we have tried with a non-changing chip and an encryption code that is duplicated 2,000 times? Won’t we have tried all 10,000 possible combinations in a matter of hours? Source: QZ

However, this method mentioned by Congressman Darrell Issa too has its limitations since the iPhone only runs code signed by Apple and a signing key is required for the brute force attempt to succeed and overcome the artificial passcode delays.

For the iPhone 5C a new version of iOS should be enough as only the OS prevents you from doing that (you don't have programmatic access to F2' without special privileges). For newer iPhones, the Secure Enclave does prevent that apparently. The FBI can't compile their own, hacked version of iOS without this restriction because the iPhone only runs code signed by Apple and the FBI does (probably) not have Apple's signing keys. Source: Crypto.SE

2

Probably

There is little disagreement that AES-256 cannot be cracked (it's still the US Gov't standard for encrypting Top Secret documents) or that the cryptographic key (UID key) cannot be extracted by software means.

However, there is whole range of physical, invasive techniques with which it is feasible (albeit difficult) to extract UID key from the microchip. Presentation "Physical Attacks on Tamper Resistance: Progress and Lessons" by Dr Sergei Skorobogatov (University of Cambridge) describes these techniques and claims following results for similar microchip ("military use Actel ProASIC3 secure FPGA family"):

How long does it take to get the AES key?

  • Initial evaluation time for all attacks from 1 week – 1 month
  • Invasive attacks (microprobing)
    • 1 day with FIB and probing station
  • Semi-invasive attacks (side-channel and fault attacks)
    • 1 week/1 hour with optical emission analysis (FDTC2009)
    • 1 hour with optical fault injection attack (CHES2002)
  • Non-invasive attacks (side-channel attacks)
    • 1 day with low-cost DPA setup: resistor in VCC core supply line, oscilloscope with active probe and PC with MatLab software
    • 1 hour/10 minutes with commercial DPA tools (DPA Workstation from Cryptography Research Inc. or Inspector SCA from Riscure)
    • 1 second with QVL-E board using special SCA sensor from QVL
    • 0.01 second with Espial tester using breakthrough approach to power analysis technique from QV

One of the above techniques has been used in 2010 by hacker to crack TPM chip.

No doubt FBI has resources and access to experts to use any range of above techniques. Of course they might be reluctant do use these techniques, as invasive techniques have high risk of destroying the hardware irreversibly before data can be extracted.

-1

A lot of information is available publicly from Apple's website, at https://www.apple.com/business/docs/iOS_Security_Guide.pdf .

There are several ways how one might access the data on this phone, at least in theory.

Each file on the device is protected unless open by SHA-256 used as the hashing function. If you can crack SHA-256 at a reasonable cost, problem solved. I very much doubt it is possible. I very, very, very much doubt that we would hear about it if it is possible. It is 99.9999% certain that the FBI isn't going to crack the phone this way. It is 100% certain that we will never hear that it is cracked that way. Anyway, Apple cannot help with this.

Part of the encryption process is a 256 bit key built into the CPU. The CPU can encrypt or decrypt data with that key, but it will not reveal the key itself. If that key was known then decrypting the phone would be very simple. In that case you could make a copy of the flash drive, copy it to a supercomputer, and try all possible passcodes. But the key isn't revealed by the CPU, and Apple doesn't record the keys and possibly never knows them. BUT the key is somewhere in the CPU. It might, or it might not, be possible to cut the CPU open and search for the key under the microscope. This is very difficult and expensive. It will destroy the phone, so if it doesn't work, all is lost. There is a 256 bit key. If you don't get all 256 bits that makes the decryption job a lot lot harder. I don't know if it's possible. It would depend on exactly how the CPU is built. If possible, it would mean Apple's help is not needed. Apple cannot help with that.

There was a comment that "the NSA can do difficult and expensive". Yes, they can. I don't know how difficult it would be to extract the key from the CPU. But also, if the NSA can do it, they will want to keep this very secret. They would only use this to handle some really important case. In this case, the victims are dead, the killer is dead, there is quite possibly nothing interesting on this phone at all. The NSA wouldn't make their capabilities public for this. Maybe secretly if Apple can't be ordered to help.

Without these methods, the only way to unlock the phone is to give it the right passcode. But the user set up the phone so that its contents is erased after 10 incorrect attempts. It has been reported that the FBI did 8 attempts and they failed, so they have one free attempt and one that erases the phone if it fails.

The only way to get around this is to change the software that controls the passcode checking and the erasing; that software is called firmware. To write the software, and to make it work reliably so it doesn't cause damage, requires lots of knowledge about the iPhone hardware, so I would assume that only Apple engineers and only very few of them would have the know how needed to do this (it's not a matter of cleverness, but experience with the hardware. These engineers would find it very hard to say write firmware for a Samsung phone without help from Samsung). So there Apple's help is needed; that problem could be overcome by offering huge amounts of money to these engineers. But the other problem is that an iPhone doesn't just accept any firmware, it has to be signed with a key that Apple keeps very, very secret (because if it was in the open, the iPhone would be open to any hackers). That definitely needs Apple's help. Or breaking into Apple's computers and stealing the firmwire signing key.

Note that nobody knows definitely that Apple can create this software and install it on a locked phone. A locked phone might not accept new firmware.

A totally different method, often used, is to forget about cracking the phone, and figuring out how to get data out of the phone without unlocking it. If a user has set up their phone to backup to iCloud (which most users have) then the phone will backup even when it is locked. All you need is put the phone in the range of a WiFi network that it knows. Once the backup is made, Apple can deliver the backup to law enforcement, and that is done something they have done in the past (that would be "help by Apple", but it's "help by Apple that Apple will give without complaining"). Unfortunately someone changed that user's iCloud password, allegedly ordered to do so by the FBI, and now the phone doesn't know the correct password anymore and cannot backup. Nobody knows the old iCloud password so it cannot be changed back. Nobody knows if this trick would have worked, but since the password was changed, it cannot work anymore.

A different method has been suggested: It might be possible to examine the glass on the phone, and if the same passcode was used a lot, one might be able to determine the digits of the passcode. Knowing the digits, there are 24 combinations only. Since they had 10 attempts to try the passcode, they could use 8 attempts for a 33% chance of unlocking the phone, if the four digits were right. Or gathering all kinds of information about the person and finding significant numbers and trying eight of them. The FBI might find my passcode if they had to.

Other methods: Anything I can't think off. However, I don't see any evidence for Snowden's fifth claim.

  • 2
    This is very difficult and expensive. but the NSA can do difficult and expensive. – Christian Feb 25 '16 at 6:00
  • 7
    SHA256 is a hashing algorithm. Files are not encrypted with it. – Kevin Krumwiede Feb 25 '16 at 9:12
  • 7
    @gnasher729 : No, they don't they say SHA-256 is used as the hashing function. If the difference between hashing and encryption isn't clear to you, you are not in a good position to evaluate the difficulty of the FBI breaking the key by reading the Apple documents. – Christian Feb 28 '16 at 2:27
  • 3
    Clearly you haven't heard of side-channel attacks, or even rubber-hose cryptanalysis. It is a very, very far stretch to call anything impossible in cryptography. Just because an algorithm is used doesn't mean it must be cracked to decrypt the entire file. – March Ho Mar 1 '16 at 14:57
  • 3
    -1: this answer is speculation, based on author's gut feeling ("99.9999% certain", "Other methods: Anything I can't think off"), makes elementary mistakes like not knowing the difference between SHA-256 (hashing algorithm) and AES-256 (encryption algorithm) and completely ignores existence of physical attacks such as theregister.co.uk/2010/02/17/infineon_tpm_crack – vartec Mar 1 '16 at 18:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .