Rapid growth of the internet users and their use of email on one hand and the exponential increase of unsolicited users sending spam have made the email system unreliable. There are several intelligent anti spam systems which use different AI techniques to filter out spam including Neural Networks and Fuzzy Logic systems. This paper presents a spam detection system based on interval type-2 fuzzy sets. Obtained results demonstrate the potentials of interval type-2 fuzzy set as an effective technique in spam detection and email classification. The proposed system also enables the user to have more control over the various categories of spam and permits the personalization of the spam filter.