AI, Terrorism and the New Digital Battlefield

When AI-generated images from conflict zones spread across social media within seconds, millions of people can see them before anyone verifies their authenticity. During the Israel-Hamas war, synthetic images and videos circulated faster than governments and journalists could confirm or debunk them.

This raises a critical question: if states struggle to contain AI-driven misinformation during a crisis, how will they respond when terrorist and extremist groups deliberately use the same technology?

That question is no longer hypothetical.

Extremist Groups Are Embracing AI

The UN Analytical Support and Sanctions Monitoring Team has warned that the threat from Islamic State Khorasan (ISIL-K) has increased. The report highlighted the March 2024 attack in Moscow and growing threat levels across Europe.

It also raised concerns about Tehrik-e-Taliban Pakistan (TTP) and the return of Al-Qaida personnel to Afghanistan for training and reorganization.

Technology did not create terrorism. However, it has changed how terrorist groups operate.

Large language models, chatbots, image generators, voice-cloning tools and translation systems can help extremist groups create propaganda at unprecedented speed. These tools can produce deepfake videos, synthetic voice recordings, multilingual material and deceptive online content.

AI and social media can serve legitimate purposes. The same technologies, however, can also be exploited for radicalization, recruitment and propaganda.

From Online Propaganda to AI-Assisted Operations

ISIS was among the first major terrorist organizations to exploit encrypted messaging platforms and online propaganda. The group used digital platforms to reach supporters across borders and spread its ideology.

ISIS-K, also known as ISKP, has taken that strategy further.

UN briefings to the Security Council have described the growing use of advanced digital tools by ISIS and ISIS-K. These tools can support recruitment, propaganda and the financing of cross-border attacks.

Three areas are particularly concerning: online radicalization, targeted recruitment and financial concealment.

Extremists can also use gaming platforms and social media to reach specific audiences. Cryptocurrency can make financial transactions harder to trace. AI tools can further increase the speed and scale of these activities.

Evidence of AI Use Is Growing

Recent evidence suggests that extremist groups are moving beyond basic propaganda.

In March 2026, Politico reported that ISIS-K’s English-language magazine, Voice of Khorasan, had instructed followers on using AI chatbots for research and propaganda while reducing their digital exposure.

The Center for Strategic and International Studies has also documented AI-related training within ISIS-K’s propaganda operations. The group reportedly experimented with deepfake technology to create a fabricated news bulletin after the Crocus City Hall attack.

The threat may extend beyond propaganda.

A study from the Cambridge Programme on AI Science and Policy found evidence that ISWAP, an Islamic State affiliate in West Africa, used AI to assist with operational problems. The study described AI as a potential battlefield consultant rather than simply a content-generation tool.

This represents an important change.

AI could give relatively small extremist groups access to capabilities that previously required specialist knowledge, technical expertise or significant resources.

AI Safety Gaps Could Become a Security Problem

The most worrying evidence concerns the AI systems themselves.

A benchmark released on July 1, 2026, by Tech Against Terrorism tested 27 leading AI models against almost 2,500 prompts based on real terrorist use cases.

The results revealed significant gaps.

Full refusals occurred in only 57 percent of responses. Another 15 percent involved what the organization described as “hedged compliance”. In these cases, a model initially issued a warning but then provided harmful information.

Around one-third of the responses reportedly provided useful assistance beyond what a basic web search could offer.

Tech Against Terrorism has also documented more than 30 real-world cases in which AI allegedly acted as an operational assistant in terrorism or mass violence. Those cases involved at least 11 different tools and were linked to more than 70 deaths.

These findings raise questions about how AI companies test their systems before release.

Current frontier-model evaluations often focus on cyber, chemical, biological and nuclear risks. Terrorism can receive less attention, despite the potential for AI to support extremist activities.

Why Pakistan Is Particularly Vulnerable

The issue carries direct relevance for Pakistan.

The Global Terrorism Index 2026, published by the Institute for Economics and Peace, ranked Pakistan as the world’s most terrorism-affected country for the first time in the index’s history.

Pakistan recorded 1,139 terrorism-related deaths and 1,045 incidents in 2025. The figures represented the country’s worst year since 2013.

The TTP accounted for 595 attacks and 637 deaths, according to the index. Targeted killings also increased sharply as militants expanded their use of drones.

Pakistan’s militant threat has long been connected to instability in Afghanistan. The country’s security challenges have evolved through several phases, from the Soviet invasion to the rise of the TTP and the wider War on Terror.

Digital radicalization is now adding another layer to that threat.

Extremist content on platforms such as X, Telegram and TikTok can reach vulnerable audiences quickly. Algorithms can also help harmful content spread beyond its original audience.

This creates a challenge that traditional counterterrorism frameworks were not designed to address.

Pakistan Needs a Digital Counterterrorism Strategy

Controlling social media alone will not solve the problem.

Pakistan needs stronger digital literacy programs that teach young people how to identify manipulated images, deepfakes and extremist propaganda. Schools and universities can play an important role in developing these skills.

The government should also strengthen cooperation between intelligence agencies, technology companies, researchers and civil society organizations.

AI developers have a role to play as well. Safety testing should include realistic terrorist use cases before advanced models become widely available.

International cooperation will be equally important. Extremist networks operate across borders, while digital platforms and AI companies also operate globally. No single government can address the problem alone.

The New Digital Battlefield

Artificial intelligence and social media are not threats by themselves. The danger comes from how extremist actors exploit them.

The nature of terrorism is changing. Future conflicts will not be fought only on physical battlefields. They will also unfold across social networks, encrypted platforms and AI-powered digital environments.

Traditional security measures will therefore not be enough.

Governments must combine intelligence capabilities with digital literacy, responsible AI governance, stronger platform cooperation and international coordination.

The countries that adapt early will be better prepared for the next phase of terrorism. Those that ignore the digital battlefield may find themselves responding to threats only after they have already spread.

Author’s Bio: Amna Aman  is an intern with the Centre for Counter Terrorism and Extreme Violence program at the Institute of Regional Studies (IRS), Islamabad. The views expressed in this article are solely those of the author and do not necessarily reflect the views of the Institute of Regional Studies or any affiliated organization.

Advertisement