I wonder if the nft tech could be used to mark everything real with a proof of authenticity. Then those things that aren’t on the chain would be considered suspicious and you would have automated green border around any content that is authentic and red for the outside of the blockchain
How could that help at all? Seeing as the blockchain would have no way of telling the difference between human and Ai text, and if you could find a way to automatically verify that in way way that was so efficient you could expect all the text uploaded to the internet you could just run that program locally and not be beholden to people paying a fee to post anything to the internet.
I can’t imagine any sort of verification system not being completely overrun by bots/people on fiver/ mechanical turk immediately unless you tied it to meatspace IDs in an know your customer sort of way, in which case you would definitely need a central organization to do said verification, which eliminates any possible need for a blockchain as said organization can just use a faster, far cheaper, and most importantly for this application editable database.
More to the point, no one doubts that an article published by one organization was secretly published by another, but rather that they secretly used AI in the writing process, which also negates the system because that organization is never going to tell you which articles are done by AI, and any sort of reporting system for the entire organization or a specific author is just going to be immediately and constantly used to review bomb.
Well, whatever the solution to this problem is, I’m fairly sure “put a blockchain on it” isn’t going to be it. Distributed ledgers do potentially have some uses, but using them to carry “proof of humanity” information doesn’t make much sense
that’s the point of Altman’s Worldcoin Orb. It makes a hash of the picture of your eyeball and stores the hash (not the image) so you can be authenticated as a human.
Because it’s a centralized system owned by a sociopath billionaire gathering unchanging, personal details about swaths of the population using ye olde “for the greater good” adage as the justification. You’d have to be a special kind of fool to go along with it.
So again deleted my comment accidentally omg. I very like this idea. It’s burdensome but every phone already has simillar scanner. I wonder about privacy though. Can it be anonymised like monero?
I imagine it as an optional feature and then verified people can opt in to only show other verified humans in some kind of next gen web similar to fediverse. Then two layers form naturally one of verified humans and the good old internet Wild West.
I wonder if the nft tech could be used to mark everything real with a proof of authenticity. Then those things that aren’t on the chain would be considered suspicious and you would have automated green border around any content that is authentic and red for the outside of the blockchain
How could that help at all? Seeing as the blockchain would have no way of telling the difference between human and Ai text, and if you could find a way to automatically verify that in way way that was so efficient you could expect all the text uploaded to the internet you could just run that program locally and not be beholden to people paying a fee to post anything to the internet.
I think there needs to be a verification process to get on blockchain. Some kind of reputation system maybe
Akin to certificates. Problem is poisoning though or other manipulation. It will be really fun problem to solve
I can’t imagine any sort of verification system not being completely overrun by bots/people on fiver/ mechanical turk immediately unless you tied it to meatspace IDs in an know your customer sort of way, in which case you would definitely need a central organization to do said verification, which eliminates any possible need for a blockchain as said organization can just use a faster, far cheaper, and most importantly for this application editable database.
More to the point, no one doubts that an article published by one organization was secretly published by another, but rather that they secretly used AI in the writing process, which also negates the system because that organization is never going to tell you which articles are done by AI, and any sort of reporting system for the entire organization or a specific author is just going to be immediately and constantly used to review bomb.
Yeah it is very difficult problem, fun stuff
I’d give a lot to reap the fame of the one who solves it but haha I am well below the skill level
Well, whatever the solution to this problem is, I’m fairly sure “put a blockchain on it” isn’t going to be it. Distributed ledgers do potentially have some uses, but using them to carry “proof of humanity” information doesn’t make much sense
If everyone can sign their content, who verifies them as “real”?
that’s the point of Altman’s Worldcoin Orb. It makes a hash of the picture of your eyeball and stores the hash (not the image) so you can be authenticated as a human.
everyone seems to be utterly freaked out by this.
Because it’s a centralized system owned by a sociopath billionaire gathering unchanging, personal details about swaths of the population using ye olde “for the greater good” adage as the justification. You’d have to be a special kind of fool to go along with it.
I guess blind people really aren’t human… (/s)
So again deleted my comment accidentally omg. I very like this idea. It’s burdensome but every phone already has simillar scanner. I wonder about privacy though. Can it be anonymised like monero?
I imagine it as an optional feature and then verified people can opt in to only show other verified humans in some kind of next gen web similar to fediverse. Then two layers form naturally one of verified humans and the good old internet Wild West.