This paper presents a new method for islanding detection. The proposed method depends on applying an artificial intelligence method based artificial immune system (AIS) with sandia frequency shift (SFS) method. AIS is used to improve the performance of the SFS method including the improvement of the system power quality, reduction in non-detection zone (NDZ) and reducing the islanding detection time. The proposed method is tested and applied to a test system and the results show that the proposed method is more efficient in islanding detection.