logoalt Hacker News

cgearhart • today at 3:59 PM • 0 replies • view on HN

The most "practical" thing I can point to is that the Miller primality test has a concrete bound under the generalized Riemann hypothesis (which is also claimed in the OP paper). Still not very practical because we already had the algorithm and we already knew experimentally that it worked well in practice.

This is why I compared it to "life on another planet". In both cases it might be more surprising if it was false; but even knowing the answer for certain, we wouldn't have any immediate application for it. It's likely that with further study we can learn more and might find new applications.

It is a remarkable achievement, but I suspect that in 100 years the most notable thing about the Riemann Hypothesis will be that it was among the first tremendous achievements of AI. It'll be picked apart and dissected, extended, and applied, but just as we rarely think about how silicon wafers are made into the computers we're using, future generations will not stare up at Riemann as a mountain daring them to climb. They'll see it as we look at other foundational results from the past: a stepping stone on the path to the next mountain.