Hy, after the recent news about the demand of AI data centers causing prices of pc components to increase at an absurd rate, I started wondering if we (the average pc user) can do anything about it.I'm just curious, is there a way we can attack AI's data, so AI will not be a reliable source for any subject you attack, and if so what would it be and how could we employ it.I'm just getting tired at the affects of AI, while not being able to get a say in any of the things AI and AI data centers are messing up.
There is nothing you can do. Humanity is doomed. Make your time.
A Markov chain that produces functional looking but nonsensical code blocks. You could get an existing language model to produce comments and function names, just make sure to switch them around randomly so they're wrong. Then start pushing into new Github repos. Would probably need a good botnet to simulate activity to look legit; stars, issues, PRs. In theory you can extend this to plaintext; with reddit posts, substack entries, or even released books.But all of this assumes they still scrape the usual sources for their datasets. If they only use a sanitized pre-2022 dataset, it's useless.
Lie about everything online. Don't post a factual statement ever again.
>>107725866you're falling victim to propaganda, AI isn't doing anything wrong or causing any of the current issues.
>>107725866Im also tired but there’s nothing we can do, at least legally
>>107725866The issue is that these gigacompanies hoard. If we had a higher level of decentralisation we could fight them off.So I think we need to fight by promoting local AI deployment.