Past research on intergroup contact has focussed mainly on the power of positive contact to reduce racism, while generally overlooking the potential detrimental effects of negative contact. Past research has also found, however, that people typically weight negative information more heavily than positive information. Thus we argue that negative contact should increase racism more than positive contact reduces it. In the present study we tested this hypothesis. Participants (N = 111; White Australian undergraduate students) were tested one-on-one by either a White (ingroup member) or Muslim/Middle-Eastern (outgroup member) experimenter who behaved in a positive, negative or neutral manner towards them. In line with hypotheses, participants were more aware of the experimenter’s racial group membership when the experimenter was Muslim/Middle-Eastern, than when the experimenter was White. Contrary to hypotheses, we did not find that negative contact with a Muslim/Middle-Eastern experimenter led to more racism than positive or neutral contact with a Muslim/Middle-Eastern experimenter. In fact, there was no interaction predicting racism. There was a main effect however, such that participants reported less prejudice, and they took fewer sweets that were offered as a gift from the Muslim/Middle-Eastern experimenter, compared to the White/European experimenter. Implications and future directions are discussed.