Convika

הדף עדיין לא זמין בעברית, ולכן הוא מוצג באנגלית.

What is the Riemann Hypothesis? A plain-English explanation

Quick answer

The Riemann Hypothesis says that every nontrivial zero of the Riemann zeta function has real part 1/2. Put another way, the prime numbers are spread as evenly as they possibly can be. Bernhard Riemann stated it in 1859, it is one of the 7 Clay Millennium Prize Problems with $1 million for a solution, and on November 20, 2026 Levent Alpöge, a mathematician at Anthropic, posted a proof he says he found with Claude and formalized in Lean, which mathematicians are now checking.

Publish my explainer free

Free plan: no credit card, 1 published page.

On this page

The Riemann Hypothesis is a statement about where the zeros of one function, the Riemann zeta function, lie. That sounds far from everyday numbers, but the zeros decide how the prime numbers 2, 3, 5, 7, 11, … are spread along the number line. The hypothesis says they all sit on one line, and that is the same as saying the primes are as evenly spread as they can be.

Bernhard Riemann stated it in 1859, in a paper titled “On the Number of Primes Less Than a Given Magnitude”. It has stayed open since then. On November 20, 2026, Levent Alpöge wrote on his website: “I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis.” Mathematicians are checking it: MIT mathematician Andrew Sutherland has compiled it in Lean on his own, and the full machine check is still running as of November 21, 2026, 10:50 JST. The news, and where the check stands. Every fact on this page was read on November 21, 2026 (Japan time), with the sources at the end.

How the primes are spread

A prime is a whole number above 1 that only 1 and itself divide. Primes thin out as numbers grow: 4 of the numbers up to 10 are prime, and about 1 in 14 of the numbers near a million.

The prime number theorem, proved in 1896 by Jacques Hadamard and Charles de la Vallée Poussin, says how fast they thin out. Near a number x, about 1 number in ln x is prime, where ln is the natural logarithm. Adding up those chances gives the logarithmic integral, li(x), the best simple estimate of π(x), the count of primes up to x.

The estimate is remarkably close:

xPrimes up to x, π(x)li(x) − π(x)Error as a share of π(x)
1,000 (10³)168106%
1,000,000 (10⁶)78,4981300.17%
1,000,000,000 (10⁹)50,847,5341,7010.0033%
10¹²37,607,912,01838,2630.0001%
10¹⁸24,739,954,287,740,86021,949,5550.00000009%

The error keeps growing in absolute terms but shrinks fast as a share. The prime number theorem only says the share goes to 0. The Riemann Hypothesis is the claim about exactly how small the error stays, and that is where the zeta function comes in.

The zeta function

The zeta function starts as an infinite sum:

ζ(s) = 1 + 1/2ˢ + 1/3ˢ + 1/4ˢ + 1/5ˢ + …

For s = 2 the sum is 1 + 1/4 + 1/9 + 1/16 + …, which Leonhard Euler showed equals π²/6. For s = 1 it is the harmonic series, which grows without limit.

Euler also found why this sum knows about primes. It equals a product with one factor for each prime:

ζ(s) = 1/(1 − 2⁻ˢ) × 1/(1 − 3⁻ˢ) × 1/(1 − 5⁻ˢ) × 1/(1 − 7⁻ˢ) × …

This is the Euler product. Every whole number is a product of primes in exactly one way, and multiplying out the product produces each term 1/nˢ exactly once. So any fact about ζ is a fact about the primes.

Riemann's step was to let s be a complex number, s = σ + it, with a real part σ and an imaginary part t. The sum only adds up when the real part is above 1. Riemann showed that the function has one natural extension to every complex number except s = 1. This is called analytic continuation: picture a road you can see only on one side of a hill; there is exactly one smooth way it can continue on the other side, so the visible part fixes the whole road.

Zeros and the critical line

A zero is a value of s where ζ(s) = 0. The extended zeta function has two kinds:

  • Trivial zeros at s = −2, −4, −6, and every negative even number. They are easy to find and say nothing deep.
  • Nontrivial zeros, all inside the strip where the real part is between 0 and 1, called the critical strip. They come in pairs mirrored across the vertical line where the real part is 1/2, the critical line.

The first nontrivial zero is at s = 1/2 + 14.134725…i. The next two are at 1/2 + 21.022040…i and 1/2 + 25.010858…i. Every zero found so far has real part exactly 1/2.

The Riemann Hypothesis: every nontrivial zero of ζ(s) has real part 1/2. The Clay Mathematics Institute puts it this way: “all the ‘non-obvious’ zeros of the zeta function are complex numbers with real part 1/2.”

Why the zeros control the primes

Riemann wrote an exact formula for counting primes. It is the smooth estimate li(x) plus one correction wave for each nontrivial zero. A zero with real part σ adds a wave whose size grows like x^σ.

So the zeros' real parts set the size of the error in the table above. If every real part is 1/2, every wave grows like √x, the smallest size possible, since zeros on the line exist. One zero off the line, with a real part above 1/2, would make the error larger than that from some point on: the primes would cluster more than the hypothesis allows.

Helge von Koch made this exact in 1901: the Riemann Hypothesis holds if and only if the error li(x) − π(x) stays below a constant times √x · ln x. For x = 10⁹, √x · ln x is about 655,000, and the real error is 1,701. That is why “the primes are spread as evenly as possible” and “every zero is on the critical line” are the same statement.

What is known

Mathematicians have proved a lot about the zeros without proving the whole hypothesis. The main steps:

YearWhoResult
1914G. H. HardyInfinitely many zeros lie on the critical line
1942Atle SelbergA positive proportion of the zeros lie on the line
1940sAndré WeilProved the Riemann Hypothesis for curves over finite fields, an analogue in a different number system
1974Pierre DeligneProved the analogue for all varieties over finite fields (the Weil conjectures)
1974Norman LevinsonAt least 1/3 of the zeros lie on the line
1989Brian ConreyAt least 2/5 lie on the line
2004Xavier Gourdon and Patrick DemichelThe first 10¹³ zeros (10 trillion) checked by computer, all on the line
2020Kyle Pratt, Nicolas Robles, Alexandru Zaharescu, and Dirk ZeindlerAt least 5/12 lie on the line
2021David Platt and Tim TrudgianEvery zero up to height about 3 × 10¹² checked rigorously, all on the line
2026Levent Alpöge and Ralph FurmanAt least 2/3 of the zeros are simple and on the line, posted on arXiv on August 13, 2026. The paper says the proof was “discovered autonomously by Claude”, and it comes with a Lean 4 formalization
2026Levent Alpöge, with ClaudeA proof of the full hypothesis, posted November 20, 2026 with a Lean formalization, being checked

Each proportion result counts what share of all the zeros, up to a growing height, must lie on the line. Going from 1/3 to 2/5 took 15 years. Going from 5/12 to 2/3 took 6.

What changes if it's proven

Many published theorems start with “assume the Riemann Hypothesis”. A proof turns all of them into theorems at once: the sharpest error in the prime count, bounds on the gaps between consecutive primes, and many results about primes in arithmetic progressions, which depend on the generalized version of the hypothesis.

Read the news post for what Alpöge posted and how the check is going.

Common myths

  • “A proof breaks internet encryption.” No. RSA rests on how hard it is to factor a large number into its two prime factors. The Riemann Hypothesis describes how primes are spread on average. It gives no fast way to factor a given number, and the encryption your browser uses today stays the same.
  • “Computers already proved it by checking trillions of zeros.” No. There are infinitely many zeros. Checking 10 trillion of them is strong evidence, and a proof has to cover all of them.
  • “A Lean check settles it instantly.” Lean checks that each step follows from the last. Mathematicians still read the formal statement to confirm it is the Riemann Hypothesis itself, and confirm the code builds. That review is under way for the November 2026 claim.
  • “The $1 million is paid when the proof is posted.” The Clay Institute's rules ask for publication in a qualifying outlet, at least 2 years after publication, and general acceptance by mathematicians before it considers a prize.

Make your own explainer page

A teacher, a study group, or a curious reader can ask Claude, ChatGPT, or Gemini for a one-page explainer of the Riemann Hypothesis, with the prime table, a chart of π(x) against li(x), and a picture of the critical line. Ask for “one HTML file”. Then publish it with Convika: in Claude, ask it to publish the page through the Convika connector, or upload the HTML file on Convika. The first time, click Publish free, and the page goes live at a link anyone opens on a phone. A new version goes up at the same link.

Frequently asked questions

What is the Riemann Hypothesis in simple terms?

It says the prime numbers are spread as evenly as they possibly can be. Technically, it says every nontrivial zero of the Riemann zeta function, ζ(s) = 1 + 1/2ˢ + 1/3ˢ + …, has real part exactly 1/2. Riemann's formula ties those zeros to the count of primes, so where the zeros sit decides how far the primes stray from their average spacing.

Why is the Riemann Hypothesis so hard to prove?

The statement covers infinitely many zeros, so no amount of checking proves it, and the tools that work for some of them don't reach all of them. Methods since Selberg's 1942 work prove that a fixed share of the zeros lie on the line, and the share grew slowly: 1/3 in 1974, 2/5 in 1989, 5/12 in 2020, and 2/3 in a 2026 preprint by Alpöge and Furman. Reaching 100% needs a different idea.

What is the prize for solving the Riemann Hypothesis?

$1 million from the Clay Mathematics Institute, which named it one of 7 Millennium Prize Problems on May 24, 2000 in Paris. Its rules ask for publication in a qualifying outlet, at least 2 years since publication, and general acceptance by the mathematics community. Of the 7 problems, only the Poincaré Conjecture is solved.

Has the Riemann Hypothesis been proven?

A proof was claimed on November 20, 2026 by Levent Alpöge, a mathematician at Anthropic, who says he found it working with Claude and posted it formalized in the Lean proof checker. Mathematicians are checking it now, including whether the Lean statement matches the hypothesis. The news post tracks where that check stands.

Why does the Riemann Hypothesis matter?

It gives the sharpest possible estimate of how many primes lie below any number, an error of about √x · ln x, as von Koch showed in 1901. Many results in number theory assume it, so a proof makes all of them theorems at once. Primes are also the base of RSA encryption, which is why the hypothesis gets attention outside mathematics.

What is the first zero of the zeta function?

The first nontrivial zero is s = 1/2 + 14.134725…i, on the critical line. The trivial zeros are at −2, −4, −6, and the other negative even numbers.

Does the Riemann Hypothesis affect cryptography?

Not directly. RSA's security rests on how hard it is to factor a large number, and the Riemann Hypothesis gives no method for factoring. The Miller test, which tells whether a number is prime, is proven fast if the generalized hypothesis holds.

Sources

All read on November 21, 2026 (JST); Scientific American's report and the independent check read again at 10:50 JST.

Turn your explainer into a link anyone can open

Ask Claude, ChatGPT, or Gemini for a one-page explainer with charts. Convika publishes the HTML at a lasting link, from the AI chat through the Convika connector or from the file you upload.

Publish my explainer free

Free plan: no credit card, 1 published page.