Yes, however my (Others may have other concerns, this is just off the top of my head) chief concern was the breaking a major barrier - in that explicitly user-hostile code would be running on the device itself, one I own. I’d say it’s more of the equivalent of club employees entering your home to check your ID prior to, or during your club visit, and using your restroom/eating a snack while they’re there. (scanning would use “your” device’s resources)

There’s also the trivial nature of flipping the require_iCloud_photos=“true” value to “false” whether by intention or by accident. I have an open ticket with Apple support where my Apple Maps saved locations, favorites, guides, Home, reports, reviews ALL vanished without a trace. Just got a callback today saying that engineering is aware of the problem and that it’s expected to be resolved in the next iOS update. I’m the meantime, I’m SOL, so accidents and problems can and do happen, nor is Apple the police.

And on top of that there’s also concerns of upstream perversion of the CSAM database for other purposes - after all, who can audit it to ensure it’s use for CSAM exclusively and who can add to it? Will those images from the device and database be pulled out for trials or would it be a “trust the machine, the odds of false positives are x%” situation? (I believe those questions might have been already answered when the controversy was flying but just a lot of cans of worms waiting with this.)