Electricd comments on OpenAI says it's scanning users' ChatGPT conversations and reporting content to the Police

[–] Electricd@lemmybefree.net 14 points 1 year ago (1 child)

I mean, I see this as a consequence of all these articles about people using ChatGPT for harmful things

[–] BreadstickNinja@lemmy.world 7 points 1 year ago (1 child)

Yeah, they're damned if you do, damned if you don't. Both they and Google are getting sued over kids who committed suicide, whose parents should have been monitoring them and getting them mental health treatment. If the courts decide that LLM companies bear legal and financial responsibility for user actions, then of course they're going to do this.

The only privacy is local. And actually, given Microsoft, local and Linux-based.

permalink

fedilink

source

parent

hideshow 2 child comments

[–] Electricd@lemmybefree.net 4 points 1 year ago*

Coming soon: McDonalds get sued for selling burgers to a minor that ate 3 burgers every day and died! McDonalds must set thresholds per customer and collect IDs from minors!

I would be for holding companies responsible when they fuck up, like McDonalds clearly marketing burgers to minors and saying it's healthy, but we must not hold them more accountable than they really are

The only privacy is local. And actually, given Microsoft, local and Linux-based.

Soon: huggingface gets sued because hosted models are being used for getting drug recipes, or did not actively prevent people from killing themselves through it

By the way, I was recently testing https://nano-gpt.com/ which claims to have privacy through TEE models... but I don't see how it's private in any way? It just guarantees that the output went through some TEE, but it doesn't guarantee that the input and output didn't leak elsewhere or got logged

permalink

fedilink

source

parent

lemmy:	0.19.20
mlmym (mschae):	26.3.4