Sorry for the newbie question, but isn't most internet traffic end-to-end encrypted, these days? So what information would the hackers, or for that matter the "lawful intercept" system , have been able to steal? I do see how accessing routers would let intruders launch malwares, spoof other sites for phishing attacks, etc.
https://www.cnet.com/tech/tech-industry/nsa-disguised-itself...
2. Even if they did, chrome has enforced certificate transparency, so a gag order on the CA/CT provider would simply result in the certificate being rejected.
As far as you and i know. Those things are not public. Helps with espionage (see Crypto AG).
Also this problem would apply to any key like gpg. Well, as long as it's not in a Hardware security module. Of course they could also seize that but at some point it becomes logistically impractical, at least for mass surveillance.
After the PRISM stuff, folks got a lot more savvy with encryption. TLS has been tightened up a lot since then across many fronts (perfect forward secrecy, removing crap roots, certificate transparency, etc).
There's just no way the NSA can be MITMing any reasonable proportion of traffic. Possibly extremely targeted stuff, and sure, there's technically the possibility that Google is handing over keys, but if it was happening at any massive scale, people would now know.
That's why the fight has moved over to metadata now, which is what the three letter agencies are vacuuming up these days.
If access is wide-spread, you could even figure out who's communicating with who over encrypted messengers by watching for packet timings. Target A communicates with the Signal server and milliseconds later Target B receives a push notification? And then seconds later the inverse happens? That's probably proof enough that two people are communicating.
I doubt lawful intercept systems have the ability to inject any traffic, but it's very useful to know the exact make, model, modem version, and OS version of a phone before sending malware like Pegasus to a device, and telco infrastructure knows most of that.
As for phishing, knowing what services your target uses can be very useful. Spoofing numbers isn't very hard, and if you've been receiving calls from your local real estate agent for a while, you won't notice as much when you an imposter uses a spoofed number. The more niche and offline the business you're pretending to be, the less likely you'd consider a phishing attempt suspicious.
Isn't that mitigated by certificate transparency?