With an emphasis on digital radicalization in Islamic contexts, the current study examines AI deployment tactics for combating online extremism. It looks at the potential of AI, regulatory frameworks like as the EU AI Act, and the technological adaptability of extremists. The 2023 AI propaganda manual of the Islamic State and the spread of extremists to gaming platforms are two examples. The study examines three AI models—overt analytical, direct engagement, and covert engagement—and draws attention to conflicts in the definition of “Keyboard Jihad.” The most successful AI agents for religious instruction are transparent, community-partnered ones. The study supports the maslaha (public interest) argument for visible AI participation but opposes covert AI because of ethical and legal issues. It suggests a three-pronged approach: deploying analytical AI right now, trying direct engagement agents in pilot programs, and delaying covert AI until legal approval.

https://www.preprints.org/frontend/manuscript/7b9fe7abe2bb1048c8afd044ec515991/download_pub