Abstract

This paper will analyze how and to what extent terrorists and violent extremistshave interacted with generative artificial intelligence so far and identify potentialways in which they could misuse generative AI in the future. It is therefore necessaryto question assumptions that terrorist and violent extremist actors will quickly adoptgenerative AI solely based on an assessment of the capabilities it may offer. Althoughresearch has shown that technological capability and availability are key drivers of terroristinnovation, terrorist and violent extremist actors also assess any new technologyon its compatibility (with both their modus operandi and ideology), relative complexity,cost, and the context in which they are operating. Understanding these factors, both inisolation and in relation to one another, will be a key aspect in monitoring the extent towhich terrorists and violent extremists adopt generative AI tools. In doing so, the paperwill seek to identify a likely trajectory for the abuse of this technology by terrorist actorsas well as conclude with some initial recommendations for policymakers.