https://www.apple.com/ca/newsroom/2021/11/apple-sues-nso-gro...
Also thanks to 5 eyes knowledge sharing your gov in US and my gov here don't need warrants to spy on their own citizens. They just have your data pass through a foreign but 5 eyes hosted service in one of their partner nations and pass data back via intel sharing agreements.
The only safe assumption is all your devices are compromised or can easily be.
The problem is keeping it under the radar. If NSO starts blasting their exploits to the general population it'll be a matter of days before some smart security researcher figures it out and gets it patched. Because even if the exploit itself is hard to detect, you still have the C&C and exfiltration to deal with which is detectable in many ways. You can block doing it via WiFi but some researchers even run their own cell hotspot.
I think this is why they make it so expensive to it remains really rare. It protects their business.
This is a very common problem in cryptography. Once you're discovered your target will simply change their tools. The allies had to let a lot of people die in less important convoys lest the Germans would get suspicious about Enigma being cracked. And always had to fake a 'chance' encounter with a spotter plane or something.
I thought the way this works is that NSO Group determines there's a vulnerability in ios or android that lets them insert Pegasus, but soon after it gets deployed to spy on 1 person by a NSO Group client, Apple and Google eventually detects it and patches the vulnerability.
I don't think it scales to 'all iphones'.
There is a recent Frontline episode where NSO Group denies any knowledge of how Pegasus gets deployed, once they sell the exploit system to a client (like a government), but that seems false to me since it is my understanding NSO Group runs a large amount of infrastructure to collect data from infected phones and to report data to the client, a multi tenant system.
I can't find a source for Snowden saying this either.