If $Cov(X, Y) = 0$, $X$ and $Y$ must be independent.

If $X$ and $Y$ are independent, $Cov(X, Y)$ is always $1$.

If $X$ and $Y$ are independent, $Cov(X, Y) = 0$.

Covariance measures the strength of non-linear dependence.