I like to say that science is defined not by the answers it has but by the questions it has. John Cole adds to his already impressive collection of ID debate by linking to a TNR article that touches on the necessity of another theoryfor the development of species. (Hint: None)
Allow me to offer a different formulation on the concept of falsifiability (that neologism is bound to send the spell check nuts). Theories often have competition, multiple ways of organizing present facts into a coherent framework. The question for the scientific method then becomes to extrapolate from each theory and make predicitons about the shape of future facts. Where the predictions differ offer the potential to find new data (read: experiment) that will agree with the predictions of one, the other, or neither theory.
The new data may not necessarily be a death blow to the "losing" theory. It is entirely appropriate for the proponents to take the new datum into account and slightly alter the theory. Once done, predictions are made with the new theory and another experiment can be devised. Note: the way that a theory has to fit into the environment of facts and the minor changes in theory to conform to the demands of that environment place is evolutionary in nature.
A true scientific theory would have to be proactive enough to create predictions that would fit their theory and not Evolutionary theory. By introducing an intelligence, the only way to distinguish itself would be to discover a species alteration that would survive despite being unfit to the environment. That in itself would rock evolutionary theory to its core (which is the political goal of Intelligent Design) but would not establish ID as the reigning paradigm. ID would have had to predict the mutation through communication with the intelligence, no fair pointing to the new species and saying "He meant to do that". This is the dangerous part because one would have to prove that communication with the intelligence occured in order to link the cause of the intelligence (which can be arbitrary by definition) and the effect of the new species. I'm not sure that ID proponents are ready for the presence of a scientifically probable intelligence. That can take a whole lot of umph out of the idea of faith.