4 ms·
An interesting thing in this whole tempest is the notion of the CSAM itself. Specifically, there are more than a few people whose argument against Apple seems t
by neurobashing 5y ago
An interesting thing in this whole tempest is the notion of the CSAM itself. Specifically, there are more than a few people whose argument against Apple seems to be, implicitly, that CSAM is something of a moral panic. It exists but not at the scale that Apple needs to embed this technology in every device. Other posts alleged NCMEC lies about stats, ostensibly to justify their budgets (if CSAM is not a growing problem, then they don't need more funds). Others insist that the existing means of policing don't work in an increasingly technical world; and the bad guys are good at this stuff. I don't know which side is right - or if there even is a "right", here.
To me, then, the whole thing at least partly rests on those questions. If CSAM is a moral panic, then this technology is not just bad, but doubly so. Apple got played by the Feds, or maybe they really have been dreaming of backdooring their devices and finally had the rationale they needed.
But if CSAM isn't a moral panic but an actual growing problem that needs a technical solution (or assist), what are we to do? Just shrug and say "well it's a hard problem, maybe one day we'll solve it" and hope for the best?
- freddybobs 5y agoThis article on HN today https://tutanota.com/blog/posts/wer-wird-belauscht/ https://tutanota.com/blog/posts/wer-wird-belauscht/ "Politicians regularly claim that they need to ban encryption to protect the children. But who is actually being monitored?" Apparently by orders of magnitude over any child related issue, it's drugs. That seems to imply that it is a "moral panic". If they said they were doing for drugs, or drugs + other stuff then that might be more convincing. Moreover if the article is to be believed - such mechanisms are used at hugely higher rates for things they are not pitched as being for. So that breaks the narrative of 'we'll only use it for X'.