OpenAI Unveils AI Solutions to Ten Longstanding Mathematical Problems
Artificial intelligence research firm OpenAI announced that its unreleased internal model, Astra, successfully generated solutions to ten longstanding mathematical and theoretical computer science problems that had remained unsolved for at least a decade. The discovery highlights growing computational capabilities in advanced research, though the initiative has drawn broad scientific interest alongside ongoing evaluation of AI-driven mathematical proofs.
Key points
- OpenAI, an artificial intelligence research and deployment company, revealed that its unreleased internal model named Astra solved ten long-stagnant math and computer science problems.
- The targeted problems spanned fields including high-dimensional geometry, coding theory, arithmetic circuit complexity, group theory, operator algebras, quantum complexity, lattice cryptography, and extremal combinatorics.
- The solutions were achieved using a token volume that would cost approximately $2,000 at standard Sol API rates.
- The development builds on prior milestones, such as an AI-generated disproof of the Erdős unit-distance conjecture shared in May.
- The initiative aligns with OpenAI's broader academic outreach, including a program providing free model access to 100,000 scientists and mathematicians.
OpenAI has announced that an internal version of its next major artificial intelligence model, designated as Astra, successfully generated solutions to ten longstanding problems in mathematics and theoretical computer science. According to the company, these specific problems had remained open with no progress on their main results for at least a decade, and in many cases significantly longer.
The breakthroughs span multiple complex disciplines, including high-dimensional geometry, coding theory, arithmetic circuit complexity, group theory, operator algebras, quantum complexity, lattice cryptography, and extremal combinatorics. OpenAI reported that the total number of computational tokens required to find these solutions would cost roughly $2,000 based on standard Sol API rates. Once discovered, the mathematical arguments were prepared into formal research manuscripts.
This announcement follows previous demonstrations of advanced mathematical reasoning by OpenAI systems, including an AI-generated disproof of the Erdős unit-distance conjecture revealed in May. These ongoing evaluations are part of a broader effort by the company to support the scientific community, which recently included the launch of ChatGPT for Academic Researchers to provide 100,000 scientists and mathematicians with free access to top-tier models. The latest results have already begun inspiring further academic exploration across theoretical computer science and mathematics.
Sources
The WireByte editorial team synthesises technology news from multiple primary sources, verifies the facts, and links every source. Articles are produced with AI assistance and reviewed under our editorial policy.