Last Updated on May 23, 2026 by Staff
For almost eighty years a famous math problem had people stumped. This problem was first thought of by a mathematician named Paul Erdős in 1946. It was a geometry puzzle that many people around the world tried to solve. No one could figure it out.
Now some researchers are working with intelligence and they think they may have made a big breakthrough. They used an AI model made by OpenAI and it made some major progress on this problem. This AI model is suggesting that Paul Erdős original idea might be wrong. This is exciting for mathematicians because it shows how AI can help solve problems that people could not solve on their own for a long time.
The Original Problem
The question behind this problem is actually pretty simple. Imagine you have a piece of paper. You put dots on it randomly. The problem is to figure out how many pairs of dots can be the distance apart.
Paul Erdős thought that the number of pairs that’re the same distance apart would only grow a little bit faster than the number of dots. Many mathematicians agreed with this. No one could prove it.
This problem is hard because it involves a lot of areas of math like geometry and number theory. Many smart mathematicians tried to solve it but they could not.
AI Helps With Math
The breakthrough happened when some researchers at OpenAI used one of their AI models to look at this problem. They asked the AI model if Paul Erdős’ idea was wrong.
The AI model did some math. Came up with some surprising results. It generated hundreds of pages of math reasoning using concepts from geometry and complex numbers. By putting dots on a flat piece of paper the AI model used higher-dimensional math to solve the problem.
This approach allowed the AI model to find arrangements of dots that had more pairs that were the same distance apart than people thought was possible. The research paper says that this construction is like a dimensional version of Paul Erdős classic example but it is much stronger.
The results suggest that there may not be a limit on the number of pairs that can be the same distance apart like Paul Erdős thought.
Human Mathematicians Check The Work
Even though the AI model did a lot of the math, human mathematicians were still necessary to make sure the results were correct. They checked the math line by line to make sure it was valid.
After checking the work the mathematicians agreed with the result. The proof says that there is a value that allows the number of pairs that are the same distance apart to grow much faster than people thought was possible.
This was a step because math depends on being accurate. AI models can generate results that look good. Human mathematicians have to make sure that every step is correct.
The fact that AI models and human mathematicians can work together shows how research may change in the future. AI models can help come up with ideas and human mathematicians can make sure they are correct.
Why This Discovery Matters
This breakthrough is important because it shows how AI can help with math research. For a time AI models were only used for calculations and finding patterns. In this case the AI model helped come up with new math ideas that people had not thought of before.
Many scientists think that this could change how people approach math problems. AI models can test ideas, look at many possibilities and find hidden patterns in complex equations.
At the time this raises questions about what role mathematicians will play in the future. By replacing human mathematicians AI models may become a tool that helps people come up with new ideas and discover new things.
A New Era Of Discovery
The solution to this math problem is a moment for both math and artificial intelligence. A problem that people could not solve for a time may be closer to being solved because of the work of humans and machines.
Even though there is still work to be done this breakthrough shows the potential of AI in scientific research. From geometry, to physics and medicine AI models are starting to help researchers in ways that were not possible before.
For math this could be the start of an era where AI models do not just calculate answers but help come up with new ideas.