For a random variable $X \sim \text{Exp}(\lambda)$, the score function is $U(\lambda) = \frac{\partial}{\partial \lambda} \ln L(\lambda)$. What is the Fisher Information $I(\lambda)$ defined as?