Police took my Pixel 6a and iPhone 12 pro

matchboxbananasynergy Y'all can correct me if I'm wrong here, but assuming the phone is in BFU state when the advesary gets a hold of it, and assuming that you use a password that distrusts the secure element (90+ entropy password/passphrase), I fail to see how they can get anything from it, seeing as everything is at rest and cannot be bruteforced.

I imagine from a forensics perspective it's already extremely difficult to succeed without this type of setup, so this would make it so much more. I'd still never say never and assume anything can be exploited with enough time. GrapheneOS already does a lot to make certain extraction methods not possible in comparison to Stock.

Nuttso you're one of the very few people that seem to understand what kind of importance data integrity has Infront of a court.

Thanks. While I focus on information security now I used to do Mobile forensics, hence how I am qualified for using UFED (A Ruggedized UFED Touch2 to be exact). Luckily never been to court but know many people who have.

I read the thread and the conclusions you make are about the same as mine, although I think Cellebrite are less capable than both what they describe and what people think they might have, mainly because when things get as sophisticated as this, they simply aren't for Cellebrite to deal with. I also made a post a while ago about how GrapheneOS features make Cellebrite unintuitive and why they would likely avoid using them entirely here:

https://discuss.grapheneos.org/d/4727-graykey-countermeasures/30

Oftentimes I find the overhype being just people grouping Cellebrite with intelligence-agency level threats which isn't comparable or realistic. These tools quite literally exist so investigators can do the work with minimal knowledge, time, and effort while maintaining forensic integrity. I would not be inclined to think an intelligence agency who will classify everything would give a shit about ANY integrity considering they would kill to get information on some people depending what state it is. Plus, sometimes knowing information on a target could be so good for their operations it doesn't matter if the evidence is invalid to them or not.

Nuttso I'm working with bunch of lawyers in Germany hand in hand on the encro and sky ecc cases. This is our work here:

I can't say a lot about these since my knowledge basically boils down to hyped news articles. From what I've seen these devices have always been total garbage, and the fact they could get away with making a phone that essentially relies on trusting a centralised, targeted and at-risk source to function is really bad... not to mention their insecure hardware.

I don't remember where this was said but I remember hearing that GrapheneOS gets targeted by companies who sell that crap because the existence of a real, open and secure mobile OS like this one puts these criminal markets out of business. Big reason why I donate and support this project.

Nuttso When quantum computing reaches maturity it will crack AES, RSA, etc. Also, NIST has begun planning the transition to quantum-resistant encryption algorithms.

Intelligence agencies have been hoarding data to be decrypted once their quantum capabilities are sufficient. Crazy times.

Kryptos That is not true. AES isn't really in any danger. Quantum just reduces the time needed to crack it by half. Symmetric encryption is quantum safe. Asymmetric encryption not. This is why we try really hard to get funding for molly to implement a quantum safe algorithm.

Kryptos Nuttso
For further reading, I found this paper to be really valuable:
Post-quantum cryptography - dealing with the fallout of physics success (Bernstein and Lange)

From table 1 on page 3:

Name: AES-256
Function: block cipher
Pre-quantum security level: 256
Post-quantum security level: 128 (Grover)

Name: RSA-3072
Function: encryption
Pre-quantum security level: 128
Post-quantum security level: broken (Shor)

Name: RSA-3072
Function: signature
Pre-quantum security level: 128
Post-quantum security level: broken (Shor)

Keep in mind that the definition of "post-quantum" is fuzzy and uncertain. There are many players in the academic-industrial-military complex overhyping its imminence in the pursuit of investment / grant money.

As far as I am aware, there are no reputable expert quantum physicists in the field that expect this to happen sooner than 10 years from now, but most of them expect it to happen within the 21st century.

Nuttso Symmetric encryption is quantum safe. Asymmetric encryption not. This is why we try really hard to get funding for molly to implement a quantum safe algorithm.

Thought I sent a reply about last replies, guess it didn't.. I don't think AES with a good enough key size is weak by any means, I just think:

• Science can be very unpredictable sometimes
• Most people wont go out of their way to target the encryption, but the systems that make the encryption possible. While I think this is completely miniscule of a risk for a securely designed system like Pixels, all about time, budget and interest.

I would love to see Kyber or other NIST PQ algorithms get tried out in messaging applications soon. I'm not much into the quantum/science stuff though so I can't comment.

zzz There are many players in the academic-industrial-military complex overhyping its imminence in the pursuit of investment / grant money.

As someone in there this is true. Too much obsession with commercial trends, and creating useless solutions from a lot of junk put together (eg. Blockchain technologies + something like Childcare).

V1king They are now allowed to use violence here for FaceId and fingerprint unlock.

Personally if I was at risk of violence I'd only use GrapheneOS + everything on a profile so I can delete that profile during a tough time, although I'm not at risk of either so maybe I can't think of any better plan.

zeroNULLvoid it's unfortunate that people have to break the law so we could gain equal rights for everyone on this planet. The war on drugs is a joke. The only solution is legalization of every substance. Nothing else will change what is happening now. As long as this doesn't change. Criminals will use gos or similar.

And having high ranking criminals use grapheneos is the best that can happen to grapheneos. Free forensic and penetrating testing against the OS.