Presumably you've come across this question in your four years of development and would have exact numbers (perhaps not for my chosen value of tx/sec) already at hand. The fact that we're three comments deep into this, leads me to believe you are dancing around the question.
At this point I think it's perfectly fair to start with the assumption that a new Crypto is a scam and it needs to do the legwork to show that it isn't. To claim that a blockchain (the most ludicrously inefficient data-structure ever devised) can scale to a billion users is an outrageous claim. The technical means they found overcome this problem should be front and center in their documentation.
The question becomes: what is tx throughput at N billion users? What are the scaling strategies that will get us there? It is zk-rollups (or zk-zk rollups)? Is it sharding? Is it moving to custom hardware circuits? I suspect it will be some combination of all of the above.
We don't know what the answer is yet and we will devote tremendous resources to figuring it out. I don't want to give the impression that MobileCoin as it is written today will scale to Alipay levels of tx throughput, but I do believe there is a path to get there that requires a ton of work.
Does that answer your question?