AWS often deploys its new platform products and latest hardware (think GPUs) into us-east-1, so everyone has to maintain a footprint in us-east-1 to use any of it.
So as a result, everyone keeps production in us-east-1. =)
Not to mention, even if you went through the hassle of a diverse multi-cloud deployment, there's still something in your stack that has a dependency on us-east-1, and it's probably some weird frontend javascript module that uses a floatilla of free tier lambda services to format dates.
2) People who thought that just having stuff "in the cloud" meant that it was automatically spread across regions. Hint, it's not; you have to deploy it in different regions and architect/maintain around that.
Eh, us-east-1 is the oldest AWS region and if you get some AWS old timers talking over some beers they'll point out the legacy SPOFs that still exist in us-east-1.
