Al-Qaida and the Islamic State are using artificial intelligence in ways that could strengthen propaganda, recruitment, and operational support, but experts say the two groups are doing so differently because of their distinct organizational cultures. The broader warning from researchers is that AI is not transforming terrorism overnight, but it is lowering the barriers for extremist content creation, online engagement, and some planning tasks.
How groups are using it
Experts and monitoring groups say extremist organizations are using generative AI for many of the same practical tasks: creating propaganda, localizing messages, evading censorship, and supporting recruitment. Tech Against Terrorism says hostile actors are chiefly using generative AI to speed up human content creation, rather than replace it entirely.
The U.N.-linked UNICRI office says the malicious use of AI by terrorist actors is now a recognized counterterrorism issue, and researchers are tracking how these tools can be adapted for influence operations and other abuses. That includes AI-generated images, text and other media that can make extremist messaging faster, cheaper, and more varied.
One common concern is that chatbots could be used to groom potential recruits in a conversational setting before a human recruiter takes over. Another is that AI may help produce propaganda that looks more polished, more emotional, and more targeted to specific audiences.
Al-Qaida’s slower approach
Researchers say al-Qaida tends to treat AI more deliberately, in keeping with its longer-term organizational style. According to the analysis, pro-al-Qaida outlets were circulating likely AI-generated imagery as early as 2023, and some of its networks have promoted chatbot workshops and guides.
That fits a movement that often thinks in decades rather than weeks. Instead of chasing the newest tool for immediate impact, al-Qaida-linked media and supporters appear to be exploring how AI can fit into a broader propaganda infrastructure.
Experts say that makes al-Qaida’s approach more incremental but potentially more durable. If a group builds AI into its media routines, it can normalize the technology among supporters and integrate it into recruitment, translation, and content adaptation over time.
Islamic State’s faster style
The Islamic State, by contrast, is described by analysts as moving quickly and seeking immediate effect. Observers from extremist monitoring groups say Islamic State-linked communities are more openly encouraging followers to use new digital tools to spread ideology, avoid platform moderation and recruit.
That “move fast” style is consistent with the group’s broader online culture, which has long relied on speed, spectacle, and adaptation. In practice, that means AI can be used to amplify official propaganda, produce supporting media, and flood online spaces with content designed to sustain attention.
The Islamic State’s approach also appears more decentralized. Instead of one central AI doctrine, its supporters and unofficial channels seem to experiment with tools as they become available, creating a looser but often faster-moving ecosystem of abuse.
Why experts are worried
Experts say AI does not need to be revolutionary to be dangerous. Even modest gains in speed, scale and language adaptation can help extremist groups reach more people with less effort.
That lower barrier is the key concern. Where teams once needed skilled designers, translators or editors, generative AI can now help create images, text, and multilingual variants at far lower cost.
Researchers also warn about the possibility of deepfakes, synthetic audio and manipulated media that can revive old extremist narratives or make propaganda feel newly relevant. Because AI can imitate style and tone, it may also help extremists produce content that appears more credible to unfamiliar audiences.
Limits and realities
Still, experts caution against overstating AI’s impact. The strongest assessments so far suggest AI is more of an enabler than a standalone game-changer.
Tech Against Terrorism says current terrorist use is chiefly augmentative, speeding up the human process rather than replacing it. CSIS argues AI is likely to make propaganda, recruitment, and some operational tasks easier, but not to transform terrorism overnight.
That distinction matters because it keeps the focus on risk without slipping into hype. Analysts say extremist groups still face limits in trust, access, operational security, and the challenge of translating online content into real-world action.
The counterterrorism response
Governments, platforms, and monitoring groups are trying to catch up. One line of defense is better detection and takedown of extremist media, including the use of hashes to identify known terrorist images and videos.
Another is disrupting the workflow before content spreads widely. That means watching for AI-generated propaganda, monitoring public guides that teach extremists how to use the tools and tracking how those networks adapt when platforms update moderation systems.
Some researchers also argue for using AI against extremists by generating counter-messaging and positive content. But experts caution that any such effort has to be carefully designed so it does not become a new vector for abuse.
Why this matters now
The spread of generative AI has made extremist experimentation cheaper, faster, and more scalable. That does not mean terrorist groups have suddenly become technologically dominant, but it does mean their propaganda systems can move with less friction than before.
The core concern is not one spectacular AI plot. It is the steady accumulation of smaller advantages: faster translation, sharper images, easier recruitment contact, more persuasive messaging, and quicker adaptation to platform enforcement.
That is why experts say the most realistic response is not panic, but persistence. The fight will likely be won or lost in moderation systems, digital literacy, intelligence work and the ability to keep extremist groups from turning useful technology into routine propaganda infrastructure.