I am wondering whether it is a known result whether for every natural number $a \geq 1$, there is at least one natural number $b$ (of any size) such that $a^2+b^2$ is prime.
This seems empirically certain, since
- there are infinitely many primes of the form $a^2+b^2$, and
- there seems to be at least one prime of form $a^2+b^2$ for every $a$ even with the added constraint that $a \geq b$: see this post: the answer there directed us to a sequnece in OEIS, where it is claimed that with the added constraint, this is an open problem.
Without the added constraint, I am curious whether it's known.
A related conjecture is the Bunyakovsky conjecture, but I am not asking whether there are infinitely many $a^2+b^2$ primes for a fixed $a$, which is very well known to be an open problem.
I am asking whether for every $a\in\mathbb N$, there is at least one $b$ such that $a^2+b^2\in\mathbb P$. This does not preclude the possibility of all integer-valued polynomials having at most finitely many primes, for example.
I printed out a line only when the smallest possible $b$ was strictly larger that previous $b$ values