P versus NP
The hodge conjecture
The Poincare conjecture
The Riemman Hypothesis
YangMills existence and mass gap
Navier Stokes existence and smoothness
The Byer and Swinnerton-Dyer conjecture
Of all these i'm only familiar with the first one, P versus NP, as should all Computer Science majors. I find it to be a particularly nice problem since finding a polynomial time algorithm solving the Clique Problem is enough to show P=NP granting a profit of the sum above mentioned. If, after you've tried really hard to find a polynomial-time solution to the Clique Problem, you come to think that it's not possible, then proving why it's not possible will show P<>NP, granting again the quoted sum. There's no losing with this

