Yes, I have because it isn’t.

It’s LLMs and Machine Learning, though ‘learning’ is a misnomer so probably better to call it LLMs

could? Lol.

ACAB.

The internet archive is a good idea.

In practise it has a lot of flaws like enabling harrasment and doxxing etc of individuals, as they take very little down, even if it would lead to abuse.

It’s technically decentralized, but functionally this doesn’t mean anything (yet) except for them not having to care about moderation.

It seems part of it might be open source, like the AT protocol, but not sure about all of it.

Hehe

Betteridge’s law of headlines.

So no.

Well, the web shouldn’t be human. But if they were to attempt to make it then LLMs would not be the way.

Sure, I get your point and agree.

I just think that third party middleman attacks should be mitigated wherever possible and so far, in looking at online therapy I’ve never seen one that can back up their claims of being private with data, just vague references to how secure it is.

It’s a tricky situation, and I guess there’s no good solution because either we have to trust the third party or just the therapist and if neither can be trusted, well…

it just speaks to a larger issue with society as a whole, how we treat mental health and especially that we need to pay in order to deal with our issues, and that it hasn’t been understood yet that it’s shameful and scary to come forward about problems because of certain laws, and the possibility of either the original scenario or the one you proposed.

This is terrible. However it’s one reason online ‘therapy’ should never be trusted to be private.

Unless the therapists start using actually private ways of communicating like Signal, then stuff like this will keep happening.

Yep, that’s why mastodon only allows for 72 characters maximum in passwords, I assume.

https://immich.app/cursed-knowledge

and an attack on encrypted messaging

lmao. It’s barely encrypted and what little of it that is (one-to-one DMs) have never been audited in a way where Telegram haven’t just either ignored the results or moved the goalposts.

Fairly new, yeah.

Oh, right, yeah, that’s bad.

That’s fair, one reason I stay out of big groups, though you can set it not to be displayed to anyone, so I’m not fully grasping your argument.

What do you mean by passport? It will a bit, if someone doesn’t know the phone number of an account that sent a message.

Okay, interesting. I still think my points about it stand though.

And that’s fine and good, if you want that, I’ve no problem with people getting their needs met.

I just wish they wouldn’t call it secure or private or think there will be no consequences for using it, there absolutely could be because there is no encryption in groups and bad encryption in one-to-one contexts.

Apart from the lack of moderation and refusal to comply with police etc from a distance, there isn’t much keeping those who use it safe from arrest, discrimination etc.

Okay, I’ll bite. But considering Signal has no data and very little metadata to give people, what exactly is the problem? What evidence would they have to arrest people on?

Especially now that people don’t have to share phone numbers to add a contact and can stop others from finding them via numbers.

Are you saying Signal uses bad encryption? I genuinely am not sure if this is sarcasm or genuine.

Did Signal roll their own encryption? I am unaware of this if so. Even if it is the case, it has been audited heavily, something which telegram have repeatedly either failed to do or moved the goalposts every time it has been audited. Telegram is not a secure messenger.

I think I was explaining why people could see it as bad, not that I particularly want more global moderation. Having said that, there should probably be a way to throw off people, on any platform, who actually do material harm to other individuals, such as distributing CSAM.