> “I didn’t know what the problem was—I was just doing Erdős problems as I do sometimes, giving them to the AI and seeing what it can come up with,” he says. “And it came up with what looked like a right solution.” He sent it to his occasional collaborator Kevin Barreto, a second-year undergraduate in mathematics at the University of Cambridge.
Seems like standard 23 year old behavior. You're spending $100-$200/mo on the pro subscription, and want to get your money's worth. So you burn some tokens on this legendarily hard math problem sometimes. You've seen enough wrong answers to know that this one looks interesting and pass it on to a friend that actually knows math, who is at a place where experts can recognize it as correct.
Seems like a classic example of in-expert human labeling ML output.
1. How can we be sure ChatGPT knows it's correct or not? It gives out incorrect answers to complex questions all the time. The very fact that it gave out a correct answer is worth talking about.
2. The type of human that can verify a mathematical proof is also the type of human that knows the appropriate communication channels to let every other math-human know about the proof. The math-humans will know the impact that proof has on math, and how to apply it.
How is he even posing the question and having even a vague idea of what the proof means or how to understand it?