
Is it really more efficient when they have tone of GPU sitting unused in warehouse because they don’t have datacenters to put them into while they continue buying more?

Is it really more efficient when they have tone of GPU sitting unused in warehouse because they don’t have datacenters to put them into while they continue buying more?
My thought exactly.

Fix the symptoms vs fixing the problem


Wow, thanks for bringing this up, that guy really is a piece of shit.


Whenever I see stories like this, all I can think about is how this is all a waste. It’s a waste of the resources needed for these agents to run while they don’t provide any value. It’s also a waste of time for the people that have to go through the AI slop.


That’s a totally valid use case, I have done it in the past and there are Linux distributions that are made to be used this way. Those distributions usually work by loading the whole system into ram which avoid multiples writes to the USB stick. They also usually offer some persistence mechanism so that data can be added on the USB stick.
Examples of distribution meant to be used this way: MX Linux, Puppy Linux
According to the article, her “process” is just doing multiple pass with AI, then having a human proofread to fix any “AI tells that have slipped through” and do some rewriting.
So, yeah, it is slop. Her goals are not to write books, but to make money by selling products to certain readers based on rigorous market research. She is also revealing that she use AI only when required by the platform.
To people that defend this heavy use of AI as she is, I always have the following arguments:
I just feel like this heavy use of AI can only lead to the ruin of all artistic endeavour, be it movies, games, books, musics, etc.