AI deepfakes weaponized in modern conflicts: systemic risks of digital warfare and corporate accountability gaps

Indigenous epistemologies emphasize communal knowledge-sharing as a defense against disinformation, contrasting with Western individualistic solutions like 'media literacy' training. Traditional oral cultures often have built-in mechanisms for verifying information through collective memory and intergenerational storytelling, which could inform modern digital resilience strategies. However, these systems are systematically eroded by extractive technologies and state surveillance, leaving communities more vulnerable to algorithmic warfare.

The use of fabricated narratives to justify war is not new; colonial powers routinely deployed disinformation to justify invasions (e.g., the 'weapons of mass destruction' pretext for Iraq). The 20th century saw state-sponsored propaganda evolve from radio broadcasts to digital deepfakes, with Silicon Valley corporations inheriting the role of propagandists from legacy media. The current crisis is a culmination of decades of unchecked corporate militarization of technology, enabled by deregulation and the outsourcing of state violence to private actors.

In South Asia, deepfakes have been weaponized to target religious minorities, reflecting historical patterns of communal violence fueled by elite interests. African digital rights activists have long warned about the weaponization of AI by both state and non-state actors, advocating for pan-African data sovereignty frameworks. Latin American communities have developed 'verification brigades' to counter disinformation, drawing on indigenous practices of communal truth-seeking.

Research shows that deepfakes are most effective in contexts of preexisting polarization, where they exploit cognitive biases and emotional triggers rather than factual inaccuracies. The scientific consensus highlights that detection tools lag behind generative AI capabilities, creating a 'whack-a-mole' dynamic where new forms of manipulation emerge faster than countermeasures. Studies also indicate that algorithmic amplification of disinformation is more damaging than the content itself, as it exploits social media’s engagement-driven design.

Artistic movements like deepfake satire (e.g., 'DeepTomCruise') reveal the absurdity of algorithmic manipulation, offering a form of cultural resistance. Spiritual traditions across cultures emphasize the sacredness of truth-telling, with rituals designed to cultivate discernment and integrity in the face of deception. Indigenous artists in the Pacific Northwest use carvings and songs to encode historical truths, providing a model for preserving collective memory in the digital age.

Scenario modeling suggests that by 2030, deepfakes could enable 'plausible deniability' in nuclear escalation, as AI-generated audio/video could frame false-flag attacks. The proliferation of synthetic media may lead to a 'post-truth arms race,' where states and non-state actors invest in AI-driven propaganda as a first-strike capability. Future governance must prioritize preemptive regulation of AI infrastructure, not just content moderation, to prevent systemic collapse of trust in institutions.

Marginalized communities in conflict zones (e.g., Palestinians, Uyghurs, Tigrayans) are disproportionately targeted by deepfakes to justify violence or erase their narratives. Women and LGBTQ+ activists face gendered disinformation campaigns designed to silence them, leveraging deepfakes to fabricate scandals. Grassroots organizers in the Global South have proposed community-led verification networks, but these are systematically underfunded in favor of top-down 'solutions' from Western tech firms.