💡 Sometimes the Most Interesting AI Stories Are About What Becomes Possible
A non-mathematician recently asked an experimental version of Claude to tackle the Riemann hypothesis—one of the most famous unsolved problems in mathematics, stumping researchers since 1859.
Claude did not crack it. Let us not get ahead of ourselves.
But in the process, Claude produced a new proof that made real progress on a mathematical limit researchers had been stuck on for decades. It organized dozens of smaller AI helpers, tested thousands of ideas, dug through old papers, checked its own work, and produced something human mathematicians reviewed and confirmed as meaningful.
The person guiding it, Jarred Sumner, mostly just cheered from the sidelines: "Keep going." "Believe in yourself."
There is something oddly human about that.
AI is not just about getting answers faster. It might help more people ask bigger questions—and make discoveries they never could have managed on their own. Not bad for a few pep talks and 31 million tokens.
That is our Feel-Good Friday story about AI. Happy fall forward. 🍂 |