When sampling from a population, the sample mean will always be closer to the population mean as the sample size increases.

Why is this statement wrong?

My explanation is that as sample size increases, the standard error decreases and the sampling distribution will become normal. Am I right?