Depending on the nature of the disaster, I think I would prefer it.

A ransomware incident where every "workstation" becomes encrypted, losing availability. We could re-create every workstation (in this case, Desktop in a Cloud) likely within an hour.

Compared to rebuilding every laptop? Days.

If availability of the Thin Client is lost, the DR might be as simple as a factory reset or could be to use a personal device to connect to the Desktop in the Cloud.

Then a COTS thin client could be re-shipped which has no customisation. Using Amazon for fulfillment, that would be at the persons house or office the following day. As it doesn't need any changes made, there is reduced pressure on IT support to re-onboard everyone.

Surely network access is often lost during disasters? Sure, for ransomware recovery, it might be good, but not any random disaster. A cheap Google Chromebook offers essentially the same hypothetical advantages but with much less dependency on the network, and with much less ongoing cost. I have no vested interest in this, I use Ubuntu and a Mac myself, but I get the appeal of wanting to have a locked down thin client. I'm not quite sure a device that depends on good internet, all the time is such a good idea, seeing how often there are often outages for networks and popular central services. Slack, GitHub, OpenAI, Cloudflare etc. If the device had some open API it could be used with, so it's not a brick in the event of wanting to switch services, it might be more interesting.