Author here. Thanks for taking the time to read.

I agree we’re in an interesting era where frontier research has shifted from mostly publicly funded to mostly private and it creates challenging incentive structures especially regarding externalized costs of research.

Did you have any thoughts on my argument of how public knowledge does get damaged by the proliferation of AI over time?

▲

typ 3 hours ago | parent [-]

I don't understand how knowledge, either public or private, could get damaged.

Though the income of the individuals and businesses that rely on the expertise of the knowledge would be damaged. Is that what you meant?

Edit: At this stage, the revenue made by the AI labs is almost entirely spent on the formation of fixed capital and opex. The demand is mobilizing physical resources with money. Atoms are relocated and reconfigured into compute racks. But eventually, the created productivity will perhaps make supply-elastic goods extremely cheap and abundant, while the supply-inelastic goods will be worth even more relative to the elastic ones.

After all, money is simply a token for the transmission of physical resources. It doesn't create stuff out of thin air. When new stuff is created, it just makes money cheaper, so that the banks can respond with more money to counteract it. More stuff -> deflation -> more money creation allowed to undo the deflation.

But the "exchange rates" between different goods and services will diverge. That's also why I don't think a direct money transfer like UBI would fix the problem, when it doesn't change the divergence of relative economic values of different goods. Let's say, extremely cheap software and entertainment, but unaffordable healthcare and housing. More money for everyone doesn't make limited resources available. So, what I am leaning into is some sort of Georgist policy. That could hopefully mitigate the price divergence, assuming that AI cannot make every commodity equally abundant.

▲

abalashov 3 hours ago | parent [-]

Well, lots of ways. One is some degree of model collapse, as the slop-enshittified Internet itself is ingested as training data--despite the AI companies' best efforts to prevent this, they won't be altogether successful.

But the more consequential one may be that few are motivated to contribute more training data to make Dario or Sam richer. This is already playing out in open source. People write open-source so humans can use it, in that human way that humans do, not to make Dario richer because his models will emit statistically convoluted copies of that open-source. What is my incentive to open-source something that I could commercialise today, compared to what it was before the LLM age?

(Many will say there's not much point in commercialising it, either, but to the extent that software still has commercial value, the appeal of the alternative path has greatly diminished.)

▲

nradov 3 hours ago | parent [-]

Dario and Sam are literally hiring human experts today specifically to create proprietary training data which their competitors can't access.

▲

abalashov 3 hours ago | parent [-]

Fine, but my point was that releasing open source software implies contributing to that, virtually by definition, and then to be met with a déluge of slop PRs and GH issues. Who can blame a developer for saying, "screw that?"

	▲	nradov 2 hours ago \| parent [-]
		I'm not blaming anyone. A lot of open source developers are paid to do that so presumably most of them will continue doing so if they want to keep their jobs. The major volunteer projects will probably introduce access controls to limit which users are allowed to create PRs and GH issues.