The "you can't possibly have better redundancy in your on-prem" was absolutely a sales pitch I heard to move to Microsoft 365.
The biggest site I work for, who went M365 hybrid last year, has incurred more Exchange downtime today than they had all of last year with their on-prem Exchange servers.
The service has redundancy, better than individual operators could provide on-prem. From the customer standpoint though, the service as a whole is a single point of failure: it either works or it doesn't.
It sounds like you are talking more about product reliability of M365 than you are about enterprise infrastructure reliability. M365 is a Microsoft product. M365 has no bearing on the credibility of cloud infrastructure as a whole, which is what your original comment spoke to.