ad sponsered content
What he does have is a ChatGPT Pro subscription, which gives him access to the latest large language models from OpenAI.
Only you can prevent forest fires
It’s both totally amazing, impressive, and utterly depressing.
Yup. The problem is not AI getting things wrong, it’s AI getting scarily good.
It’s both. But my beef with AI is that it makes life tasteless.
Hard math problem? Ask ChatGPT. Problem solved. Moving right along… What’s the fucking point of investing time and effort in anything in a world like that?
Yeah it’s both but its mistakes will get fewer and have a much lower impact than what it will accomplish. I totally agree that it makes life tasteless. In combination with machinery (past and future) AGI would make just about all human effort inconsequential and therefore rather pointless. Sure you could still make books, clothes, software and other things the old fashioned way, but basically nobody would ever use them. And knowing you’ll never make even a small difference in the world will kill purpose, motivation and mental health.
Subscription required. Any1 knows what the problem was?
Apparently this was the proof (I think? I have no real idea):
Is there some function 𝑓(𝑟) such that 𝑓(𝑟) →∞ as 𝑟 →∞, such that, for infinitely many 𝑛, there exist > 𝑎1,𝑎2 with 𝑎1+𝑎2>𝑛+𝑓(𝑟)log𝑛 such that 𝑎1!𝑎2! ∣𝑛!2𝑛3𝑛⋯𝑝𝑛 𝑟?
I have no idea what any of it means lol
I didn’t get asked for a subscription. Maybe clear cookies? It said it was an Erdos problem and this guy and his friend have been trying AI on a number of Erdos problems. The output needed tidying up to make it more understandable but it sounds like the whole proof was created by the AI, or at least all the important parts.
Thank you, here seams to be a blog post talking about the process.
https://www.erdosproblems.com/forum/thread/blog:2
I feel it is good to try to get through some of these low-hanging fruit problems for mathematicians to better spend their time thinking about the actual harder problems worth attention. I conducted this as a scientific experiment to see how far the models had progressed and what was now in reach for them.
Thanks for the link
Soon: AI solves physics problem that will enable cheap nuclear weapons
Plot twist : we already know how
Yes! Finally!!
