I don't think it can ultimately be escaped but the cited Vegh et al exactly proposes that, the bioinspiration, as a means to surpass those limitations.
However, in this article I contend that those limitations have posed little adversity in the field given the success of the latest models. As a result, it may be a bit premature to be concerned about it.
For some industries such as chemistry and biotech there is little change to be expected.
The self-employed is now able to rapidly develop their ideas while employed reducing their risk and cost.
Per VC, that's right and secondarily it's more difficult to justify earlier funding when one no longer needs to hire a substantial team. I expect the current major VC firms to get even bigger.
I try to address this in the article. I think the same could have been expected of plumbing, electricians, landscaping, etc but through a different means, standardization, the corporations simply don't have a good position.
In this world the neurodivergent is empowered, they no longer are charged with persuasion of their ideas to a team or corporation. They can build their ideas themselves and form a limited partnership with someone with the talents they lack.
Though I make it a point to describe the new economic order without regard to its desirability. However, other authors on the subject (e.g. Lysander Spooner, Rothbard, etc) would be pleased by the development in terms of its social welfare.
Again this piece is not written with GPT, feel free to ask any GPT. Ironically, maybe I should have to increase the appeal of my ideas. I chose my words carefully to communicate my ideas precisely.
We would see neither squirrels nor crows since these criticisms miss the forest for the trees. But we can address them.
> This is irrelevant for AI, because people throw more hardware at bigger problems
GAI is a fixed problem which is Solomonoff Induction. Further Amdahl's law is a limitation on neither software nor a super computer.
Both inference and training rely on parallelization, LLM inference has multiple serialization points per layer. Vegh et al 2019 quantifies how Amdahl's law limits success in neural networks[1]. He further states:
"A general misconception (introduced by successors of Amdahl) is to assume that Amdahl’s law is valid for software only". It would apply to a neural network as it does equally to the problem of self-driving cars.
> These two sentences contradict each other
There is no contradiction only a misunderstanding of what "eviscerates" means and even with that incorrect definition resulting in your threshold test, it still remains applicable.
ZIZEK: that AI will be the death of learning & so on; to this, I say NO! My student brings me their essay, which has been written by AI, & I plug it into my grading AI, & we are free! While the 'learning' happens, our superego satisfied, we are free now to learn whatever we want
Sure however the vast majority of costs come from the government regulation in the first place. Consumers of course bear the cost, however this doesn't necessarily mean we should remove these expensive regulations, one can argue they are beneficial.
If we assume they are beneficial, this also doesn't mean the government should subsidize it less but instead more to continue to allow the production of drugs. If we artificially require a price, many of these drugs would simply not be developed.
I'm under the persuasion that the future of healthcare is a voucher system to allow the competitive bidding of drugs/insurance to remain as well as the maintenance of the quality and quantity of medical research. Many countries purport to have better medical systems however we find that the American payers subsidize the rest of the world––with the USA producing more medical research than the rest of the world combined.
In a single healthcare system, the prices of drugs deemed essential will skyrocket while others will simply be excluded by the bureau. This then creates a cost on the tax payers through the inflation/excess tax required by the national budget to afford it. While at the same time the single healthcare system has little incentive to improve its efficiency of administration until it's fully deteriorated and governmental intervention promises a revamp which may still be ineffective.
The voucher system allows many insurance companies to exist, allowing the quality of insurance provided treatments to remain through that competition on their varied administrative strategies. While at the same time consumers with varying degrees of medicinal concerns can choose the right programme for themselves––opting to even pay beyond the sum provided by the voucher. This way it's no longer only affordable through a corporate benefit and the buyer group is available instead through a national benefit.
I agree, however the author is convinced about current popular solar panel technology.
Indeed even theoretically we can do 86% multi and 99% with quantum dots——in the current state of technology these remain science fiction considering the commercial manufacturing processes available.
Sure there are many novel solar technologies that subvert the SQ assumptions thereby avoiding the conclusion. However, the author of the parent post believes that existing technology today is sufficient.
You can give https://intrgr.com a try. It standardizes the Article format (removing clutter), gives you related articles from around the web, and recommendations if you make an account.
I made it because the web has become unreadable for me.