I fear we're long past the point of no return. We are exactly one 'policy update' away from not being able to install non-compliant messengers on our phones. Sure, there are still some devices that will let you unlock the bootloader and you can still sideload unverified apps, but let's be honest, most people today barely manage to install an app from the store. If installing a decentralized messenger is more involved than that, 99% of people aren't going to do it.

those platforms will be banned. this will doom Signal, the fediverse, and countless smaller platforms. anything that isn't compatible will become illegal.

I'd personally like to have a FOSS, privacy-aware CSAM (or even generic gore/porn) detector I could plug into Matrix/Lemmy/Mastodon servers. something self-hostable, so I could run those services without worrying about pedos and trolls ruining my platform.

I'm not sure if something like it exists. I'm not sure if it could exist. PhotoDNA (the old CSAM detector) ended up being somewhat reversible, so that you could actually turn signatures back into obscene material. because of this, the signature databases were shared under strict NDA, only to large players.

probably the most realistic solution is a generic porn classifier convnet. if it blocks adult porn, it should block CSAM too (hopefully?)

they are not reliant on image hashes, and reversibility concerns apply less because the dataset used to train it was presumably legal (if distasteful.)

individuals mostly have a few things they care about, but most people don't understand shit

especially technology

especially information technology

politicians are selected for being people-oriented therefore most are hopelessly underinformed

and it's very very very easy to get caught up in ideologies

and then means to an end seems like business as usual

Even if a system doesn't look authoritarian, corruption happens all the time. Those involved in corruption naturally want more power for themselves. Additionally some people actively thirst for more power for whatever reasons, and most people don't want to be constrained in their jobs, and they are all aligned in expanding governmental power. You need some discipline to commit to the idea that "I don't want the ability to see encrypted chats, even if that makes my job 90% easier to do", and I don't trust most people to have it.

Why?

Democracy is being served. People want this stuff. HN people maybe not, but let's not pretend we represent anything but a noisy minority. It's entirely democratic AFAICT, "think of the children" and "think of the terrorists" won the argument some time ago.

Free speech is not held as an absolute in many countries as it is in the US, and never has been. It is in a bad state in some places (the UK seems to be performing poorly on this measure) but a lot of places feel the right to spew whatever toxic lies spring to mind without consequences might not be entirely healthy either.

Justice? I think justice is performed better in many European nations than most of the rest of the world.

And freedom... lots of people like to claim the US is the most free place on earth, but it's really not clear that's true. There are freedoms in other countries not enjoyed in the US. Here in Australia for example, I am free to collect the rainwater and filter it for my drinking-water supply, something that's not true in every state.

Self-host may be the new Samizdat.

so you encrypt your image, send it to the government, the government runs its CSAM detector on the encrypted image and gets... an encrypted result.

then what? you decrypt the answer and send it back to them? promise that you totally didn't change the answer?

FHE is the wrong tool for the job. you'd want verifiable computation (e.g. ni-ZKP) instead. both are too complex and faaar too computationally expensive for actual use.

It would be useless because predators won't run the homomorphic encryption.

>Apple, for instance, already solved this with NeuraHash

"Solved"