Artificial Intelligence and Terrorist Content European Policy Options and Accuracy Safeguards for Arabic
This paper proposes that Europe treat accurate decisions on Arabic content as part of its counterterrorism capacity, and use artificial intelligence to assist triage and analysis under clear human responsibility. The priority is to enforce the law against specifically identified terrorist content while protecting legitimate journalism, research, religious expression and political speech. It recommends independent tests of context and dialects, prompt correction of erroneous removal, and reporting that captures both wrongful removal and wrongful retention. The appearance of AI-generated propaganda is not evidence of an independent effect on recruitment or attacks. Establishing such an effect requires causal evidence that this desk review does not provide.
