In this paper, the mean square exponential stability of a class of stochastic Hopfield neural networks with mixed delays is investigated. Employing Ito's formula and applying the inequality techniques, a sufficient condition ensuring mean square exponential stability is proved. The main result derived in this paper generalizes the results in Wan and Sun (2005), Sun and Cao (2007) and Zhu and Cao (2014) etc. An example is also given to illustrate the main result.