Let X and Y be dependent random variables. We consider the problem of designing a scalar quantizer for Y to maximize the mutual information between its output and X. We study fundamental properties and bounds for this form of quantization, which is connected to the log-loss distortion criterion. Our main focus is the regime of low I(X;Y), where we show that for a binary X, there always exist an M-level quantizer attaining mutual information of $Omega(-Mcdot I(X;Y)/log(I(X;Y))$ and that there exists pairs of X,Y for which the mutual information attained by any M-level quantizer is $mathcal{O}(-Mcdot I(X;Y)/log(I(X;Y)))$.