The Riemann Hypothesis: The Greatest Unsolved Problem in Mathematics

The Problem That Has Outlasted Everything

In 1900, David Hilbert presented his famous list of 23 unsolved problems that he believed would shape mathematics in the 20th century. The Riemann Hypothesis was number 8. It was not solved in the 20th century. In 2000, the Clay Mathematics Institute listed seven Millennium Prize Problems, each carrying a $1 million reward. The Riemann Hypothesis is one of them. It is the only problem to appear on both lists.

Bernhard Riemann stated it in 1859 in an eight-page paper — the only paper he ever wrote on number theory. The paper is considered one of the most important in the history of mathematics. The conjecture at its heart has resisted every attack for 165 years, survived the scrutiny of some of the greatest mathematical minds in history, and remains today exactly where Riemann left it: unproved.

The Cast of Characters: Primes and the Zeta Function

Prime numbers — integers greater than 1 divisible only by themselves and 1 — are the atoms of arithmetic. Every integer factors uniquely into primes. But the primes themselves appear in a pattern that is, on the surface, maddeningly irregular:

$2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, \ldots$

Gaps of 2 (twin primes), then a gap of 4, then 2 again, then 4… no obvious period, no simple formula. Yet zoomed out, primes become statistically regular.

The Prime Number Theorem, proved in 1896 independently by Hadamard and de la Vallée Poussin, states:

$\pi(x) \sim \frac{x}{\ln x}$

where $\pi(x)$ is the number of primes up to $x$ . Among the first million integers, roughly $1/\ln(1{,}000{,}000) \approx 1/13.8$ are prime. The approximation improves as $x$ grows.

But how close is “approximately”? How large is the error? That is the question the Riemann Hypothesis answers.

The bridge between primes and the answer is the Riemann zeta function:

$\zeta(s) = \sum_{n=1}^{\infty} \frac{1}{n^s} = 1 + \frac{1}{2^s} + \frac{1}{3^s} + \frac{1}{4^s} + \cdots$

For real $s > 1$ , this series converges. For $s = 2$ : $\pi^2/6$ (the Basel problem, solved by Euler). For $s = 1$ : the harmonic series, which diverges.

The Euler Product: Primes Hidden in a Sum

Euler discovered a remarkable connection between the zeta function and the primes:

$\zeta(s) = \prod_{p \text{ prime}} \frac{1}{1 - p^{-s}}$

The sum over all integers equals a product over all primes. This identity — a consequence of unique prime factorization — is one of the deepest connections in mathematics. It means the zeta function encodes information about every prime simultaneously.

When $\zeta(s) = 0$ , the product on the right must vanish — but individual factors $1/(1-p^{-s})$ never vanish. The zeros of $\zeta$ arise from delicate cancellations in the sum, and these cancellations are intimately tied to how the primes are distributed.

Extending to the Complex Plane

As a function of real $s > 1$ , the zeta function is well-behaved. The remarkable step Riemann took was to consider $s$ as a complex number $s = \sigma + it$ , where $\sigma$ and $t$ are real.

The series $\sum 1/n^s$ converges for $\text{Re}(s) > 1$ . Through a technique called analytic continuation, Riemann extended $\zeta(s)$ to a function defined on all complex numbers except $s = 1$ (where it has a pole — a singularity).

The extended function satisfies the functional equation:

$\zeta(s) = 2^s \pi^{s-1} \sin\!\left(\frac{\pi s}{2}\right) \Gamma(1-s) \zeta(1-s)$

This equation connects the value at $s$ to the value at $1 - s$ , reflecting the function across the vertical line $\text{Re}(s) = 1/2$ .

The zeros of the extended zeta function fall into two types:

Trivial zeros: at $s = -2, -4, -6, -8, \ldots$ (negative even integers). These are well understood, arising from the $\sin(\pi s / 2)$ factor in the functional equation.

Non-trivial zeros: complex numbers $s$ with $0 < \text{Re}(s) < 1$ where $\zeta(s) = 0$ . The functional equation forces them to come in pairs: if $\sigma + it$ is a zero, so is $1 - \sigma + it$ . The known zeros lie at:

$s = \frac{1}{2} + 14.134i, \quad \frac{1}{2} + 21.022i, \quad \frac{1}{2} + 25.011i, \ldots$

All of them, so far, have real part exactly $1/2$ .

The Hypothesis

Riemann’s conjecture, stated almost in passing in his 1859 paper:

It would be very desirable to have a rigorous proof of this proposition; however, I have left this research aside for the time being after some fleeting futile attempts, since it appears to be unnecessary for the immediate purpose of my investigation.

The proposition: all non-trivial zeros of the zeta function have real part equal to 1/2.

They all lie on the critical line $\text{Re}(s) = 1/2$ in the complex plane.

That is the Riemann Hypothesis. One sentence. 165 years unresolved.

Why the Zeros Control the Primes

The connection between zeros and primes is made explicit through an explicit formula Riemann derived:

$\pi(x) = \text{Li}(x) - \sum_{\rho} \text{Li}(x^\rho) - \ln 2 + \int_x^\infty \frac{dt}{t(t^2-1)\ln t}$

Here $\text{Li}(x) = \int_2^x dt/\ln t$ is the logarithmic integral (the best approximation to $\pi(x)$ ), and the sum is over all non-trivial zeros $\rho$ .

Each zero $\rho = \sigma + it$ contributes an oscillating term $\text{Li}(x^\rho)$ to the count of primes. The real part $\sigma$ determines how large this oscillation grows:

If $\sigma = 1/2$ : the oscillation grows like $\sqrt{x}$ , giving an error term $|\pi(x) - \text{Li}(x)| = O(\sqrt{x} \ln x)$
If some $\sigma > 1/2$ : the oscillation grows faster, giving a larger error

The Riemann Hypothesis is equivalent to the statement that primes are distributed as regularly as possible — the error in the Prime Number Theorem is no larger than $\sqrt{x}$ times a logarithmic factor. This is the tightest possible error bound.

Explicitly: the Riemann Hypothesis is equivalent to

$|\pi(x) - \text{Li}(x)| \leq \frac{1}{8\pi} \sqrt{x} \ln x \qquad \text{for all } x \geq 2657$

The Evidence

Despite 165 years without a proof, the evidence for the Riemann Hypothesis is overwhelming:

Computational verification: David Broadhurst, the LMFDB collaboration, and others have computed over $10^{13}$ (ten trillion) non-trivial zeros. Every single one lies exactly on the critical line $\text{Re}(s) = 1/2$ , to every decimal place of precision computed. Not a single exception has been found.

Hardy’s theorem (1914): Infinitely many zeros lie on the critical line. (Not all — infinitely many is far weaker — but a significant step.)

Levinson’s theorem (1974): At least 1/3 of all non-trivial zeros lie on the critical line. This was improved to over 40% by subsequent work.

Random matrix theory: The statistical distribution of gaps between zeros on the critical line matches, with stunning precision, the distribution of eigenvalues of large random matrices from the Gaussian Unitary Ensemble (GUE). This connection — discovered numerically in the 1970s and deepened theoretically since — suggests deep structural reasons for the zeros to lie on the critical line.

Conditional theorems: Hundreds of results in number theory, algebra, and cryptography have been proved under the assumption that the Riemann Hypothesis is true. The structure of the mathematical world that assumes it holds together beautifully.

Analogues and Generalizations

The Riemann Hypothesis is not an isolated conjecture — it is the central case of a family:

Generalized Riemann Hypothesis (GRH): The same statement for Dirichlet L-functions, which are generalizations of the zeta function attached to arithmetic progressions. GRH implies, for instance, that primes are equidistributed in arithmetic progressions in the sharpest possible sense.

Weil conjectures (proved by Deligne, 1974): Analogues of the Riemann Hypothesis for zeta functions of curves over finite fields. Weil proved the analogue for curves in 1948; Deligne extended it to varieties of all dimensions. These proofs used the full machinery of étale cohomology and are considered among the greatest achievements of 20th-century mathematics.

The Weil conjectures were proved. Their proof required inventing entirely new mathematical frameworks. Many mathematicians hope the classical Riemann Hypothesis will require similarly new ideas — which may explain why it has resisted known techniques.

Approaches and Obstacles

Over the decades, many serious attempts have been made:

Hilbert-Pólya conjecture: Perhaps zeros correspond to eigenvalues of some self-adjoint operator. If such an operator existed, the zeros would automatically be real (since eigenvalues of self-adjoint operators are real), placing them on the critical line. No such operator has been found — but the random matrix connection strongly suggests something like it exists.

Algebraic approaches: Alain Connes reformulated the Riemann Hypothesis using non-commutative geometry, connecting it to an operator on a specific Hilbert space. The reformulation is elegant but has not yet yielded a proof.

Analytic approaches: Countless attempts to bound the zeta function and its zeros using classical analysis. Progress has been made — better zero-free regions, sharper zero density estimates — but the hypothesis remains out of reach.

The difficulty is not lack of effort or talent. The problem has attracted the sustained attention of some of the most powerful mathematical minds for 165 years. The most likely explanation: the Riemann Hypothesis requires genuinely new mathematics that does not yet exist.

Conclusion: The Question That Defines an Era

The Riemann Hypothesis sits at the intersection of analysis, number theory, algebra, and physics. Its proof — if it comes — will almost certainly require ideas from fields that have not yet been connected to it.

What makes it the greatest unsolved problem in mathematics? Not just its age, its resistance, or its prize money. It is its centrality. The zeros of the zeta function govern the distribution of primes — and the primes govern the multiplicative structure of all integers. A deep truth about those zeros would be a deep truth about arithmetic itself.

Riemann sensed something in 1859. He wrote it down, called it “very desirable to prove,” and moved on. A century and a half later, we are still trying to catch up.

Related topics: Prime Number Theorem, Riemann zeta function, analytic number theory, Millennium Prize Problems, random matrix theory

Frequently asked

What does the Riemann Hypothesis actually claim?

The Riemann Hypothesis states that all non-trivial zeros of the Riemann zeta function have real part equal to 1/2. In other words, when ζ(s) = 0 for a complex number s (excluding the trivial zeros at negative even integers), then s must lie on the vertical line Re(s) = 1/2 in the complex plane. This line is called the critical line.

Why does the Riemann Hypothesis matter for prime numbers?

The Riemann zeta function encodes the distribution of prime numbers. The Prime Number Theorem tells us that primes thin out roughly as 1/ln(x). The Riemann Hypothesis would tell us precisely how closely actual primes follow this prediction — the zeros of the zeta function control the error term. A proof would give the sharpest possible understanding of prime distribution.

Has anyone made progress on the Riemann Hypothesis?

Significant partial progress exists. Over 10 trillion non-trivial zeros have been computed and all lie exactly on the critical line. The hypothesis is known to be true for 'most' zeros in a precise statistical sense. Hardy proved in 1914 that infinitely many zeros lie on the critical line. But no proof that all zeros do — which is what the hypothesis requires — has been found.

What would a proof of the Riemann Hypothesis imply?

A proof would immediately establish the best possible error bounds in the Prime Number Theorem, resolve hundreds of theorems currently proved 'assuming the Riemann Hypothesis,' and likely introduce powerful new techniques in analytic number theory and complex analysis. A disproof — finding a zero off the critical line — would be equally revolutionary, invalidating decades of conditional results.