Anthropic employee claims AI solved 150-year-old math problem
A mathematician using Anthropic's Claude AI claims to have produced a 17-million-line proof for one of the most famous unsolved problems in history.
The short version
- The Riemann Hypothesis concerns how prime numbers are distributed as they approach infinity.
- The proof was developed using Lean, a programming language designed to formally verify mathematical theorems.
- The claim comes just after Anthropic's record-breaking $81 billion stock market debut.
- The Clay Mathematics Institute has not yet officially recognized the solution as valid.
Why it matters
If verified, this would represent a massive leap in human understanding of mathematics. It also signals a shift where AI may begin to solve complex scientific problems that have eluded human researchers for centuries.
A 150-year-old mystery
The Riemann Hypothesis is one of the most famous unsolved problems in the history of mathematics. First proposed in 1859, the problem focuses on the distribution of prime numbers, which are integers divisible only by one and themselves. As these numbers approach infinity, mathematicians have struggled to find a definitive rule for their pattern.
Because of its importance, the Clay Mathematics Institute designated the Riemann Hypothesis as one of seven Millennium Prize Problems in 2000. The institute offered a $1 million prize for any person who could provide a formal solution. For over 150 years, the problem has remained a major hurdle for the global mathematical community.
Levent Alpöge, a member of the technical staff at Anthropic, recently claimed to have solved this puzzle. He announced his discovery on his personal website after working with the company's AI tool, Claude. This marks the second time in recent months that an AI company has claimed a breakthrough on a renowned mathematical problem.
How the AI proof works
Alpöge did not just write a traditional essay or a standard paper to explain his findings. Instead, he published the proof in the form of 17 million lines of Lean code. Lean is a specialized programming language that is used to formally verify mathematical theorems, meaning it checks every logical step to ensure it is correct.
The process involved a deep collaboration between the human mathematician and the AI. Alpöge noted that while he guided the work, Claude eventually outpaced him near the finish line.
The path to verification
While the claim is significant, experts note that the verification of a formal proof is a rigorous process. Alex Kontorovich, a professor at Rutgers University, explained that satisfying a computer's strict logic is a high bar for accuracy. He noted that because the computer is extraordinarily picky, satisfying it gives a very good chance of satisfying a human expert as well.
However, the transition from a computer-verified proof to human understanding is not immediate. Kontorovich emphasized that the journey from this announcement to a place of real understanding will be long and difficult. He noted that a researcher must still do the work to follow every line and figure out why the logic holds true.
A shift in mathematical research
This development follows a similar trend in the field of artificial intelligence and science. In September, OpenAI reported that its technology played a key role in a breakthrough involving the Navier-Stokes equation, which describes how fluids move. These two instances suggest a growing pattern of AI assisting in solving complex physical and mathematical equations.
The potential impact of these breakthroughs is substantial for the scientific community. Martin Bridson, president of the Clay Mathematics Institute, described the announcement as having historic significance. He suggested it holds the promise of a dramatic transformation in how people perceive what is possible when trying to extend human understanding of mathematics.
Context and corporate growth
The timing of this claim coincides with a major milestone for Anthropic on the financial stage. The company made a significant debut on Wall Street this week, raising a record amount of capital at a massive valuation. This move occurred just as the company was preparing for its major public event, adding a layer of corporate momentum to the technical claim.
Alpöge admitted that he is currently presenting a complete argument even though he has not yet mastered the massive neighborhood of mathematics surrounding it. He noted that his work with the AI produced more than 250,000 pages of material. This scale of output highlights the massive amount of data and logic the AI is now capable of generating for complex research.
Sources and links
Read the original at The Business Standard
First reported by The Business Standard. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.
Liked it? Say why
Add a line for other readers.
No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.