The correct reading of a confidence interval

X is a normal $(\mu,\sigma^2)$ random variable and a 95% confidence interval for $\mu$ based on a random sample of size $n$ on X is given by $(15,25)$. Hence it can be interpreted that: