Then they didn't have a correct backup to begin with; for high profile organizations like that, they need to practice outages and data recovery as routine.
...in an ideal world anyway, in practice I've never seen a disaster recovery training. I've had fire drills plenty of times though.
We’ve had Byzantine crypto key solutions since at least 2007 when I was evaluating one for code signing for commercial airplanes. You could put an access key on k:n smart cards, so that you could extract it from one piece of hardware to put on another, or you could put the actual key on the cards so burning down the data center only lost you the key if you locked half the card holders in before setting it on fire.
Rendering security concepts in hardware always adds a new set of concerns. Which Shamir spent a considerable part of his later career testing and documenting. If you look at side channel attacks you will find his name in the author lists quite frequently.
Not that they should just go all in on it, but an encrypted copy on S3 or GCS would seem really useful right about now.