Not to downplay what proton mail is doing, but they're saying that you can run this locally with a 2 core, 4 thread CPU from 2017 (the i3 7100, which is a 7000 series processor), and a RTX 2060, a GPU that was never considered high end. Perhaps they changed the requirements while you weren't looking. Or Am I reading this wrong?
ShakingMyHead
joined 2 months ago
I'm sure they will thank us once we explain that the alternative was GPT-5.
60% of the time, it works 100% of the time.
Probably would have been easier when the context window wasn't 128k.
Though what the point would be should someone actually achieve that eludes me a bit.
That's the problem.
Maybe I'm just reading the room wrong, but the consensus of the comment section seems to be "haha he thinks AI can replace psychics" but all psychics do is take what people said, reframe it and maybe add some nonsense after that can sound vaguely correct. For the most part, that is LLMs. LLMs could replace the psychic industry because unlike other professions there's zero obligation to be correct about anything.
Is it unethical? Absolutely. But psychics are already frauds so it's not like a legitimate profession is being replaced.
Of course, I could just be misreading the room.
TBF what's the difference between a "AI" seance and a "real" one? It's not like one is less of a grift than the other.
Took me a moment.
Unfortunately, "extremely expensive" and "high-end" aren't really synonyms, thanks to, y'know, bitcoin. Of course, I don't disagree with your argument that having to buy a GPU just to ensure your webmail does what it's advertised to do is, well, dumb.
What I don't know is what the LLM even is. Did they just tack on Llama to their webmail app and call it a day? Did they train a model? Was it trained on emails? If so, whose emails? What an advertisement that would be: "Use Protonmail to encrypt your emails so that companies like Protonmail can't use them to train an LLM."