Sorry, that was Claude 3.7, not ChatGPT 4o

https://github.com/elder-plinius/CL4R1T4S/blob/d9a004b5a29395675c5a548acfc386459f71cd14/ANTHROPIC/Claude_Sonnet_3.7_New.txt#L92

IIRC if you take s look at 4o leaked instruction (prompt that is “injected” at the begining of the chat), that model is clearly ordered HOW to solve this kind of problem lol

Your though process is wrong. You need to backpedal it. One bad apple doesn’t mean we should burn the orchard. And you can perverse anything good into an abomination.

eat the rich

How you want to do that without unions? Dude, saying out loud

[I wonder] if any union is worth it if that’s the risk of what they could become.

Is not helping anyone.

Take an L and do better.

With that mindset, why bother living or leaving the bed? Imagine the worst day you could have, why risk it?

Nah, most likely AI made the summary and that’s why it’s wrong :)

The article says its OpenAi model, not Facebooks?

And I generally don’t need 5-in-1 shower products

Well, technically a bar of soap is all-in-one product. You can even wash clothes with it.

But I get your point, I see overpriced male-targetted 3-in-1 soap gels everywhere.

It must be sad having so little manliness that smell could take from it.

Honestly, you’re not his parent, he’s an adult and needs to start taking care of himself - or GTFO.

My point is “summarizing over all of those” and “poisoning”.

Source of category 1 says cheese is made from XYZ and yellow

Source from category 2 confirms 1 in different words and adds that it has holes

Source from category 3 confirms 2 and adds that its also blue, not only yellow

Source 4 talks about blue cheese only

Poisoning would mean that in the summary cheese is yellow with blue holes.

There was literally no random Germans.

If you read memoirs from that time, or survivors letters, they always talk about Nazis being the milkman, the cobbler, the shopkeeper, the neighbour, and not Gebels or Hitler.

So scrapping “popular websites” plus “someone said this is a good source for topic X” plus wikipedia? And summarizing over them all? That sounds like a very bad idea, because it’s very fragile to poisoning?

How did we stop Nazis in the past? Was it… kindness? Memes? Or violence?

I’m not sure what’s that you’re trying to say OP.

Maybe it’s a language specific thing? In my native tongue March 23 sounds like a journal entry, not a normal date.

I get your point, but your comparison is a little… off. Wasn’t Gutenberg “printing”, not “writing”?

I stand corrected.

The AGI, by definition, will make something vastly better than diffusion model. That’s one of the cornerstones of AGI, it will explode it’s capabilities.

I think they did more - I vaguely remember them admitting they messed up the search result on purpose so the user will run more queries and will see even more ads.

Agi and image diffusion has literally nothing in common though?

Size and noise I guess.

She wore an itsy bitsy tinni minni yellow polka dot bikini 🎵