Weaponizing Gender: AI and the Future of Conflict Narratives in Southeast Asia
SRIc Insights By Htay Su Wai
Key Takeaways:
AI’s greatest gender-related risk in Southeast Asia is its ability to amplify narratives that have historically been used to mobilise violence and silence women.
Myanmar, the Philippines, and Indonesia demonstrate how AI is entering digital environments where gender is already weaponised for conflict, political competition, and social control.
Current AI governance frameworks pay insufficient attention to the intersection of AI, conflict, and Women, Peace, and Security (WPS) concerns.
Introduction
Artificial intelligence has quickly become a central component of Southeast Asia’s digital future. Governments across the region are investing in AI infrastructure, developing national AI strategies, and positioning emerging technologies as engines of economic growth. Singapore, Indonesia, Malaysia, Thailand, and Vietnam have all identified AI as a strategic priority, while regional organisations increasingly discuss AI governance in terms of innovation, competitiveness, and digital transformation (see here).
At the same time, concerns regarding the societal impacts of AI have expanded beyond technical questions of privacy and algorithmic bias. International organisations, researchers, and civil society groups have warned that AI systems can reinforce existing inequalities, including gender inequalities. Much of this discussion has focused on representation, fairness, and inclusion within AI systems. While these concerns are important, they overlook a critical dimension of the Southeast Asian context.
Across the region, AI is not entering politically neutral environments. Instead, it is being introduced into information ecosystems shaped by conflict, polarisation, disinformation, and increasingly sophisticated forms of digital political competition (see here). In these environments, gender is not simply a social category affected by technological change. Rather, gender often functions as a political instrument through which actors mobilize support, construct threats, and undermine opponents.
This distinction matters because the most significant risk associated with AI may not be that it creates entirely new forms of gendered harm. Rather, AI may amplify existing practices through which gender has long been weaponized within conflict and political struggles. For instance, Myanmar, the Philippines, and Indonesia illustrate how this process is already unfolding across Southeast Asia.
Myanmar: When Gender Becomes a Conflict Narrative
Myanmar demonstrates that gender can play a central role in how conflict narratives are constructed and legitimised. Much of the international discussion surrounding Myanmar’s information environment has focused on the role of Facebook in facilitating anti-Rohingya hate speech and misinformation. Investigations conducted by the United Nations Independent International Fact-Finding Mission found that social media platforms contributed to the spread of inflammatory content and failed to adequately moderate dangerous speech in local languages. The mission concluded that Facebook had become a significant vehicle for the dissemination of hate speech and incitement against the Rohingya population. However, the role of gender within these narratives has received comparatively less attention.
Research examining digital disinformation in Myanmar has shown that rumours concerning violence against women frequently circulated online during periods of escalating communal tension. Stories portraying Muslim men as threats to Buddhist women became powerful mobilising tools within broader anti-Rohingya campaigns (see here). Allegations of sexual violence and threats against women were repeatedly invoked to frame violence as a form of protection rather than persecution. At the same time, gender-based discrimination also exists within the Rohingya community itself. Despite facing genocide and seeking justice, some members of the community have targeted Rohingya women activists through harassment and gender-based attacks( see here).
Artificial intelligence is often perceived as a neutral technology, yet its outputs are shaped by the data on which it is trained. Because AI systems learn from existing online content, they can reproduce and amplify societal, political, and gender biases embedded in that data. In conflict settings, these biases can reinforce harmful narratives rather than challenge them. As one interviewee cited in regional research on AI and conflict observed, rumours circulating on social media regarding violence against women often preceded violence on the ground. By the time claims could be verified or disproven, communal tensions had already escalated, and attacks had begun. The significance of this case extends beyond misinformation. It demonstrates how women’s bodies, safety, and social status can become symbolic battlegrounds within conflict. Gendered narratives were not peripheral to the conflict; they were part of the mechanism through which hostility was mobilised.
This experience raises important questions about the future role of generative AI. If recommendation algorithms helped amplify inflammatory rumours during the Rohingya crisis, generative AI may significantly increase the speed and scale at which similar narratives can be produced. AI-generated text, images, and videos can now be created at minimal cost and distributed across multiple platforms within minutes. In highly polarised environments, synthetic content does not need to be entirely believable to be effective. It only needs to reinforce existing fears and prejudices. The concern, therefore, is not simply that AI will spread misinformation. It is possible that AI may accelerate the production of gendered narratives that have historically contributed to violence.
The Philippines: From Disinformation to Political Intimidation
While Myanmar illustrates how gendered narratives can contribute to communal conflict, the Philippines demonstrates how they can be used to discipline and silence political opponents. Over the past decade, the Philippines has become one of the world’s most studied cases of digital political manipulation. Researchers have documented the emergence of organised online influence networks that combine political messaging, coordinated trolling, and algorithmic amplification to shape public discourse ( see here).
Women have often been among the primary targets. The experience of Nobel Peace Prize laureate Maria Ressa and the journalists of Rappler provides one of the clearest examples (see here). Reporting by Rappler and independent researchers documented sustained online harassment campaigns involving misogynistic abuse, threats of sexual violence, and coordinated attacks intended to undermine credibility. Rather than focusing on professional performance or political positions, many attacks relied on gendered language designed to portray women as immoral, emotional, or untrustworthy. Importantly, these campaigns were not random expressions of online hostility. They functioned as political tools. The objective was not merely to insult individual women but to discourage criticism, weaken independent journalism, and shape public discourse ( see here).
Generative AI has the potential to transform these tactics. Political actors no longer need large teams to generate content continuously. AI systems can produce thousands of variations of narratives, social media posts, images, and videos targeted at specific audiences. Synthetic audio and video technologies further increase the possibility of reputational attacks based on fabricated evidence. In this sense, AI may not create a new form of political intimidation. Rather, it may dramatically reduce the costs associated with existing practices. What once required coordinated networks of operators can increasingly be automated, personalized, and scaled.
Indonesia: The Normalisation of Digital Gender-Based Violence
Indonesia highlights a third dimension of the challenge: the normalisation of digital gender-based violence. Organisations such as Komnas Perempuan and SAFEnet have documented a steady increase in online gender-based violence over the past decade. Women activists, journalists, academics, and human rights defenders frequently report harassment, doxxing, image-based abuse, and threats intended to discourage public participation ( see here). Similar patterns have also emerged among Rohingya communities, where some women activists have reportedly been targeted by members of their own community through online harassment, including the circulation of AI-generated images intended to discredit, intimidate, and silence their advocacy (see here). These cases demonstrate how AI-enabled abuse can reinforce existing gender inequalities and restrict women’s participation in civic and public life.
These patterns did not emerge with digital technologies alone. They are rooted in broader social and gender inequalities that continue to shape women’s participation in public life. Despite significant progress in women’s rights, Indonesian women continue to face discrimination linked to patriarchal norms, restrictive gender expectations, and unequal access to political and economic opportunities. Women who speak publicly on political, religious, or social issues are often subjected to heightened scrutiny and attacks on their morality, sexuality, or family roles rather than their professional qualifications or policy positions (see here). Researchers have noted that gendered stereotypes remain deeply embedded in both political discourse and everyday social interactions, creating an environment in which women are more vulnerable to intimidation and reputational attacks see here).
Unlike in Myanmar, these incidents are not primarily associated with armed conflict. Unlike in the Philippines, they are not always directly linked to organized political campaigns. Instead, they reveal how technology-facilitated gendered abuse has become embedded within everyday digital life. This trend is significant because it creates fertile ground for AI-enabled harm.
The growing availability of generative AI tools lowers the technical barriers required to create manipulated images, synthetic videos, and fabricated narratives. Activities that once required specialised expertise can now be carried out using publicly accessible applications. As a result, the production of harmful content is becoming increasingly democratised. The challenge is therefore not simply that AI creates new risks. It is that AI allows existing forms of abuse to be replicated more quickly, more cheaply, and by a wider range of actors. For women already navigating hostile online environments, this may further increase the costs associated with public participation.
Conclusion
The experiences of Myanmar, the Philippines, and Indonesia demonstrate that AI is entering information environments where gender has long been used as a political tool. In Myanmar, gendered narratives helped mobilize violence; in the Philippines, they became instruments of political intimidation; and in Indonesia, they underpin increasingly normalized forms of digital abuse. Generative AI does not create these dynamics, but it enables them to be produced, distributed, and amplified at an unprecedented scale.
As Southeast Asian governments continue to develop AI governance frameworks, the challenge extends beyond questions of innovation and ethics. AI must also be understood through the lenses of conflict prevention, digital security, and the Women, Peace and Security agenda. Without such an approach, AI risks becoming a force multiplier for existing patterns of gendered harm and exclusion across the region.
Policy Recommendations
Integrate gender and conflict perspectives into AI governance to assess how AI may amplify disinformation, digital violence, and exclusion in fragile environments.
Strengthen local-language monitoring and response mechanisms to detect and counter AI-enabled gendered disinformation and online abuse.
Myanmar already suffered the first-ever social media-fueled genocide, and to prevent AI-fueled genocide or another mass crime, the UN or international organisations need to act to manage the AI regulation in the polycrisis context
Htay Su Wai is a Junior Research Fellow at the Sustainability Lab of the Shwetaungthagathu Reform Initiative Centre (SRIc) and holds a Master of Public Policy (MPP) from the Hertie School of Governance in Berlin, Germany.
“Advocating Sustainability, Shaping Our Future”
Help Sustain The Sabai - Myanmar’s Voice for Sustainability Support The Sabai