41 ms·

>People don't need self-help advise, they need a fair redistribution of increased productivity. The increased productivity is pretty much entirely coming from

>People don't need self-help advise, they need a fair redistribution of increased productivity. The increased productivity is pretty much entirely coming from AI researchers and the companies investing in huge amounts of GPUs, and they are the ones receiving most of the windfalls, how's that not fair?

thn-gap 7mo ago
The biggest contribution is still from the training set, whose original authors get 0 because of "fair use" in the copyright.
- Noumenon72 7mo ago
  This presumes that the value was created by the authors and not the people who found a way to use the structure in the training set to create intelligence. Like crediting wi-fi router owners with creating Wi-Fi Positioning System instead of the people who realized they could wardrive around to create maps and extract a new kind of value that would never have existed without them. Or Egyptians for deciphering hieroglyphics instead of the Frenchman who realized the Rosetta Stone they were using to hold up a wall could actually be used to do much more. My thinking here is coming from the paper "From Entropy to Epiplexity"[1] which partly discussed why you can train on synthetic data: it's the structure of the data that enables learning, not just the amount of "information". Authors of images and videos may have worked just as hard as authors of text training sets, but they didn't contribute to AI as much because there just wasn't the same kind of structure to discover there. It's the people who found the usable structure, not the people who accidentally generated it, who created the value. 1. https://arxiv.org/html/2601.03220v2 https://arxiv.org/html/2601.03220v2
  - philipwhiuk 7mo ago
    > This presumes that the value was created by the authors and not the people who found a way to use the structure in the training set to create intelligence. This presumes it's binary. > Or Egyptians for deciphering hieroglyphics instead of the Frenchman who realized the Rosetta Stone they were using to hold up a wall could actually be used to do much more. Yes, I think a lot more than 1 person deserves credit for years of painstaking research. > It's the people who found the usable structure, not the people who accidentally generated it, who created the value. You're naively assuming the structure is accidental.
keybored 7mo ago
Even if that was true. There’s the training data.
oytis 7mo ago
Depends on your ethics I guess? In my view fair distribution is an equitable one, which is also the principle that helps to keep society stable. Not that researchers and business people should not be rewarded for their achievement, but the reward should be moderate and limited in time instead of creating a permanent underclass or whatever the plan is now IMO

thn-gap 7mo ago

The biggest contribution is still from the training set, whose original authors get 0 because of "fair use" in the copyright.

Noumenon72 7mo ago
This presumes that the value was created by the authors and not the people who found a way to use the structure in the training set to create intelligence. Like crediting wi-fi router owners with creating Wi-Fi Positioning System instead of the people who realized they could wardrive around to create maps and extract a new kind of value that would never have existed without them. Or Egyptians for deciphering hieroglyphics instead of the Frenchman who realized the Rosetta Stone they were using to hold up a wall could actually be used to do much more. My thinking here is coming from the paper "From Entropy to Epiplexity"[1] which partly discussed why you can train on synthetic data: it's the structure of the data that enables learning, not just the amount of "information". Authors of images and videos may have worked just as hard as authors of text training sets, but they didn't contribute to AI as much because there just wasn't the same kind of structure to discover there. It's the people who found the usable structure, not the people who accidentally generated it, who created the value. 1. https://arxiv.org/html/2601.03220v2 https://arxiv.org/html/2601.03220v2
- philipwhiuk 7mo ago
  > This presumes that the value was created by the authors and not the people who found a way to use the structure in the training set to create intelligence. This presumes it's binary. > Or Egyptians for deciphering hieroglyphics instead of the Frenchman who realized the Rosetta Stone they were using to hold up a wall could actually be used to do much more. Yes, I think a lot more than 1 person deserves credit for years of painstaking research. > It's the people who found the usable structure, not the people who accidentally generated it, who created the value. You're naively assuming the structure is accidental.

keybored 7mo ago

Even if that was true. There’s the training data.

oytis 7mo ago

Depends on your ethics I guess? In my view fair distribution is an equitable one, which is also the principle that helps to keep society stable. Not that researchers and business people should not be rewarded for their achievement, but the reward should be moderate and limited in time instead of creating a permanent underclass or whatever the plan is now IMO