However, the domain of counter-terrorism presents fundamentally new challenges for data mining algorithms — challenges that require new algorithms and new theoretical foundations for the field. Fortunately, new research is finding ways to address these challenges and, in some cases, to exploit new opportunities offered by counter-terrorism applications. I estimate it will take three to five years before we have robust commercial software that uses these new technologies.

Finally, and perhaps most important, the emerging technologies for data mining provide an array of design options that have profound differences in their potential uses and social impacts. If we understand these options at an early stage, we can select those technologies that offer the most desirable combination of uses and impacts.