Those advocating for it have a visceral pitch: your kids will be in danger if this doesn't pass. Those against are arguing for abstracts, like freedom and privacy. And sure, people should be in favour of those things - but there's a visceral pitch against, and it's not being used
What will happen if this stuff is put in place, and used as claimed? There will be a massive rush of automated "accusations" of child abuse against innocent parents, grandparents, and other relatives. This will result in many families being unnecessarily stressed and disrupted by investigations, and at least some children being removed from innocent families.
Why will this happen? after all, won't the scanning be 100% accurate?</s>
Obviously not. Tech companies don't want to be responsible for this - the scanning won't be optimised for accuracy, it will be optimised to pass the buck as hard as possible[1], because the tech companies don't want to business of taking the blame when some pedophile isn't caught.
Who is the second level of review? Maybe some minimum wage zero-hours subcontractor at Serco or Group 4 Security, or whoever the equivalent of that is in the EU. But that's probably it - once the image has been labelled 'bad' no-one else will want to look at it.
So, even if you are sending a picture of your own child to your own mother, you will have at the very least to have to think about whether it could be mistaken for child abuse by someone you don't know who has about 5 seconds to look at it and is probably from a completely different culture.
Yet none of this has been brought up in the media either by tech companies or by privacy activists.
[1] The other option is that the tech companies abdicate completely and just use some black box from the government to scan every picture. Problems with that left as an exercise...