OpenAI and Google DeepMind recently announced that their new reasoning-focused AI models solved a geometric puzzle that remained uncracked since 1946. The problem, known as the Erdős unit distance conjecture, asks for the maximum number of times the same distance can occur between points on a flat surface.
While mathematicians spent 80 years certain of a specific solution, the AI discovered a "counterexample" that proved the existing human belief was factually wrong. This breakthrough is significant because it marks the first time AI has performed original, high-level research in pure mathematics rather than just summarizing known information.
Experts are now debating whether these tools will eventually replace junior researchers or serve as a collaborative "super-calculator" for the scientific world.
.Be the first to reply to this general discussion.
Join in on more popular conversations.
