The article also mentions using linked device access. Clever. Fuck the police, though.
post
The burglar broke in by knocking on the door and waiting for someone to come to the door and open it and then asking if they could come in and were let in and asked if they could just take stuff and the person said sure go nuts.
The headline makes people think signal is somehow broken.
It’s not. Just be careful and monitor your account. And don’t let anyone gain physical control of your device.
Interesting they highlight Signal again as though this is a vulnerability.
If someone else has access to a linked device... that's you fucking up access controls.
To me, this proves that the police can still get the information they need without us handing over our encryption keys and requiring 'service providers' to MITM for them.
Except they had to get the keys, at least for Signal, as described in the article. Only you can allow a new device. If you get a notification for a new device and you go "sure, let me flash that code for you", you're giving the key. And a moron.
Can't say about the other services.
Yeah, what I'm saying is that this proves there's no need for chat control. Law enforcement can already get what they want by being sneaky.
There's also no need for chat control because there is almost always enough information anyway, but the police doesn't act on it. Their incompetency and lack of resources won't be helped by adding even more information to sift through.
for signal, that is still required. if you look, they used linked devices to get the messages
Signal failed to prevent soneone from accessing my unlocked phone and starting Signal! Everything was right there!
On iOS you can set an app to require additional credentials to open, to prevent this situation. I'd imagine Android had something similar. I did it for all my important apps, just in case. Don't want someone able to access my bank account or nudes.
If someone can get unlocked access to your phone, your security practices are already insufficient. If you really want to prevent something like this, you need to make this first step impossible, not add a bandaid on top of it.
What's described in the article is the same method Russia was using to compromise Ukrainian Signal accounts. It's just phishing.
I think you're the only person here so far who read the linked article.
I also found the netzpolitik article that was referenced but not linked to in the original article.
This describes police adding a linked device to a person's WhatsApp account while gathering evidence (translated with Google Translate).
On January 12, 2020, Mr. and Mrs. P. were questioned. During the questioning, they voluntarily handed over the mobile phones they were carrying to the interviewing officers for a brief period so that messages from their daughter contained on the devices could be viewed and, among other things, photographed.
While the photographs were being taken, the computer-based application WhatsApp Web was covertly activated via a website made available online by the Federal Criminal Police Office (BKA), allowing the messages to be read on a BKA computer (sic!). This did not involve any intrusion via a Trojan horse or similar software.
The only link to Signal here is the nation state campaign which used social engineering via a phishing message from 'Signal Support'.
This nation state campaign was originally reported by a researcher at Citizen Lab.
I did notice some comments that may be from people who did not read and understand the article. These attacks are not sophisticated or hard to defend against, which is probably why the German police want to limit public awareness.
I find their view shortsighted/narrow; the attacks they're using (including SIM swapping) are also widely used by criminals for fraud and identity theft. Police agencies should not discourage the public from being more cybersecurity-aware.
Seems like signal could send a notification 24 hours after any new device is added to remind the user that it was done. Make it so it has to be dismissed on each device so dismissing it on one doesn’t make it vanish on the rest.
Open source FIDO2 keys, KeePassXC, and Aegis.
SMS 2fa has always been a bad deal
I'll never understand why people accept SMS 2FA as any kind of security. Might as well put it as an ad in a newspaper. 🤦🏻♂️
Because many services only have SMS as 2FA option. Especially government services.
Also it is impossible to use google without enabling the SMS 2FA option. No matter what, with only 2FA authenticator app or email, they will lock down the account by saying "unable to verify".
Signal doesn’t offer anything except sms 2fa and requires a phone number. It’s a terrible choice considering LEO can do what they did here and just get legal access to MITM your sms messages, spoof the 2fa, and take over your account, impersonating you. The other ones aren’t encrypted. So none of these they broke into are great choices for truly secure messaging.
Is everyone here really whining about social engineering? It's basic comp sec, weakest link are the people.
A problem many aren't aware of, in an area that shares telecommunications info, two people within that area can be identified by the state sending encrypted messages by seeing when one person sends and another instantly receives a message. It could be easy enough to obscure that I would think by working differing lag times in.
I think it was in the intercept a few years back. I think this is it.
https://theintercept.com/2024/05/22/whatsapp-security-vulnerability-meta-israel-palestine/
Signal in this was clickbait they literally just say oh well if someone can link in their device they can see 45 days of message history
Telegram and whatsapp never had encryption. Also - they just give your messages on law enforcement request, always have.
Signal - how does it work with signal again?
They are a bit vague on this but I suspect all of these attack vectors start with LEOs having physical access to the unlocked phone. They then set up a trusted desktop without the phone owners knowing.
Which is clever, to be fair. Whether or not that's legal is already a court case. The law is so frightfully grey.
Its also a failure of the user's access control and operating security.
Once a third party has access to the secure environment, that environment is and will always be compromised.
top 50 comments