I examine different forms of the Precautionary Principle (PP) to see if these are suitable, inter alia, for the regulation of new technologies. Weak versions of the PP may be suitable, but are not importantly different from Cost-Benefit Analysis (CBA). Strong versions of the PP are importantly distinct from CBA but are not a suitable basis for regulation because they lead to paradoxical outcomes if applied consistently. I consider three different lines of response to the change of paradox and argue that all three are unsatisfactory. First, I argue against Sandin's (2007) suggestion that we should be optimistic about finding a solution to the paradox on the grounds that the PP appears to be embodied in common sense reasoning. Second, I consider Weckert and Moor's (2006) attempt to resolve the PP paradox by appealing to the distinction between positive, negative and intermediate duties. I argue that this does nothing to resolve the PP paradox in many crucial cases. Furthermore, even when it can be used to resolve the paradox, it does not provide a satisfactory resolution. Third, I argue that Gardiner's (2006) attempt to recast the PP as a form of maximin is unsatisfactory because, although it resolves the PP paradox, it can only be successfully applied in a range of cases which is much narrower than the range in which advocates of strong versions of the PP typically attempt to apply the PP.