This is an interesting perspective, however, I think this still falls short in a few areas where users have a reasonable expectation of privacy and security. Ju

This is an interesting perspective, however, I think this still falls short in a few areas where users have a reasonable expectation of privacy and security. Just a note on background, I spent over 9 years in cellular telecom, so take these biases as a long term industry insider. Apple/Google have made great progress in several areas, but there are just pieces of the puzzle we entrust the carriers with: General Location Information: By the nature of the way cellular networks work, they require all times the device is powered the relative location of the device. There are lots of rules and carrier preferences around this, but when the radio is active, the cellular network needs to know which are the closest towers, and when the radio is inactive, a wider location (think a postal/zip code). I think most people would reasonably consider leaking this information or allowing it to be public unreasonable. Specific Location Information: So I'm not totally sure about this one, since I unfortunately never took the opportunity to test it. There exists a diameter API for e911 services, that allows requesting the GPS coordinates of a device. I never tested this out, to see if the device would notify if this API was used, or whether it was only functional form within a 911 call. So take it with a grain of salt, but embedded within this might be the possibility to continually request specific GPS coordinates of a device. Denial of Service: A large issue with many of these weaknesses, is it can lead to targeted denial of service. So if I have access to the diameter network, I can send routing updates to a location where you aren't, and continually deny you access to the network. The device might be secure, but if you can't use it, it's still a problem. This wouldn't require much sophistication, probably on a similar level to say running a ddos attack against a website. Inflated Billing: Another side effect, I can target you and generate inflated bills, by indicating you are roaming on an expensive network. When I left, I believe the most expensive roaming location was still $65 CDN/MB. Try and deal with a carrier explaining that you can't possibly be in Canada and Africa at the same time. Detectability: These problems can be hard to detect, if you suspect fraud on you're account due to denial of service or inflated billing, there are only a select few people at the carrier that can find these problems. Diameter protocol also has a specific weakness, in that requests are routed by destination network, but answers are routed back along the path the request took, and may cross 3-4 companies networks. So even if you try and implement source based policy, it's trivially easy to spoof the source when implementing these types of attacks. Sim Updates: I'm not totally familiar with this part of the architecture, but if your able to spoof the device into connecting to a rogue network, you may be able to do more nefarious things than just route or block the user plane traffic. You may actually be able to send sim updates, that could brick the device or maybe even run programs on the sim card. Don't hold me to this though, I'm pretty far removed from this part of the spec. I'm also not totally convinced the internet model has all the answers yet, but I'm happy to see the progress made over the last several years. But I think that's really a different topic, so won't dive into it here. I think my argument would be progress need to be made on both sides of the equation. Carriers should work towards being able to operate with the least privilege necessary (good luck... I don't have my hopes up seeing the inside), and still protect the privacy and integrity of the information it does need (routing locations, e911 services, record keeping, OTA updates, metadata).