A Campaign without a General
As artificial intelligence further develops on social-media platforms, we can imagine heteromated terrorism being exponentially more difficult to combat than anything we have fought to this day. That is because the extremist signal will not need to originate from a particular group or state. Instead, autonomous AI bots will learn from the already existing cesspool of extremist propaganda on the Internet and build upon it independently of human direction. For states fighting terror, there will be no targets to bomb or countries to sanction. The ever-evolving enemy will live online.
As bots become more advanced at propagating extremism, they will also lack the normal constraints that even the most radical humans accept, such as the need to maintain a reputation, attend to offline commitments, or keep some semblance of peace with friends and family. Heteromated terror thus reaffirms the age-old realization that machines can do many things better than humans—and in this case, even the act of turning people into monsters. Humans who are compelled to join a violent movement in the near future may do so unaware they have done so at the behest of artificial intelligence. Under this model, acts of terror will become little more than human events that complete a non-human network’s task cycle.
Nathan González Mendelejis is assistant professor of National Security Studies at the eSchool of Graduate Professional Military Education, U.S. Air Force, and a Security Fellow with Truman National Security Project. The views and opinions expressed in this article are those of the author and do not necessarily reflect the official policy or position of the U.S. Air Force or Department of Defense.