AI video generation tools like OpenAI's Sora are no longer just creative software; they have become instruments of information warfare that governments and militant groups deploy to fabricate evidence, manipulate public opinion, and undermine trust in authentic documentation during active conflicts. From deepfake videos of political leaders surrendering to fabricated footage of military strikes, these tools are accelerating the spread of disinformation at a pace that outstrips the ability of journalists, fact-checkers, and authorities to verify what is real.

How Is Sora Being Used in Geopolitical Conflicts?

Russia has emerged as an early adopter of AI video generation for information operations. In early 2026, OpenAI reported that a Russian network used ChatGPT and the image generator Sora to run a content farm targeting Western audiences. The Kremlin has been automating its troll farms, harnessing large language models to generate human-seeming personas that work to undermine support for Ukraine on social media. Ukrainians have contended with a steady flow of AI-generated TikTok videos showing nonexistent scenes of Ukrainian troops surrendering.

The India-Pakistan conflict in 2025 demonstrated the scale of AI video weaponization. Within hours of the first Indian strike, a hyperrealistic deepfake of Pakistani Prime Minister Shehbaz Sharif went viral, depicting him "surrendering" and appearing to tell Parliament that Pakistan stood alone, abandoned by its allies. AI-generated videos falsely showed that the Pakistan Army shot down Indian aircraft, while both sides circulated fake satellite imagery showing destroyed military installations that were either intact or only slightly damaged.

Beyond video, deepfake audio has been deployed to sow military confusion. During the same conflict, deepfake audio mimicked senior Pakistani commanders issuing nuclear-alert and stand-down orders, stirring confusion among the civilian population and lower military ranks. The sophistication of these operations reflects how quickly AI tools have moved from novelty to operational weapon.

Why Can't Fact-Checkers Keep Up With AI-Generated Content?

The fundamental problem is one of speed and scale. Detection and attribution of AI-generated content run on different clocks than the spread of disinformation, and both lag the lie itself. When fabricated footage of a blast at Iran's Evin prison hit social media within minutes of a strike in February 2026, it was shared by major outlets including Sky News and the BBC before verification could occur. Forensic experts determined the video to be AI-generated within days, but by then the doubt had already done its work.

The challenge is compounded by what legal scholars call the "liar's dividend." This concept, coined in 2019 as a theoretical warning, has become the operational reality of modern conflict. The fake no longer has to survive scrutiny; it only needs to exist long enough for the truth to arrive damaged. Iran's ruling regime exploited this principle during its January 2026 crackdown on protesters, dismissing real video documentation of atrocities as fake by pointing to digitally altered versions of the same images that anyone could have posted online, creating what analysts described as an "epistemic fog" thick enough to obscure atrocities.

Steps to Strengthen AI Content Verification and Trust

• Implement Provenance Standards: Experts recommend that provenance information travel with images and videos wherever they are viewed, not remaining optional or confined to a single platform. This cryptographic signing at the moment of capture, similar to what Google's Pixel 10 smartphone introduced, would create an unbreakable chain of custody for authentic content.
• Build Detection Into Hardware: Apple's newest iPhone 18 Pro models enable users to produce cryptographically signed reference images and identify content generated or edited with AI, demonstrating that hardware-level solutions can help restore trust in the visual record before content spreads widely.
• Require Transparent Detection Tools: When California's AI transparency law took effect in August 2026 for generative AI companies, detection tools were supposed to accompany it, but testing revealed that seven of 13 AI providers had no public detection tool, and all but one of those that did could be fooled.
• Test Guardrails Before Release: Rather than retrofitting safety measures after harm occurs, experts argue that guardrails for AI tools should be tested in high-stakes environments like wars and crises before release, with some capabilities failing that test outright and being excluded entirely.

Hany Farid, the world's leading deepfake expert, noted the significance of hardware-level solutions: "When one of the largest companies in the world decides this problem is worth solving at the hardware level, it is an acknowledgment that our collective loss of trust in the visual record is no longer a niche concern for forensic researchers. It is a mainstream problem, and it demands mainstream solutions".

What Happened When Google Integrated AI Into Earth?

In July 2026, Google launched a feature that built image generation directly into Google Earth, a platform that for two decades had been synonymous with satellite image truth and was the foundation for open-source investigator training in conflict zones like Syria, Yemen, and Sudan. Users could zoom in to real coordinates, type a prompt, and get an AI-generated photorealistic scene blended into actual satellite and aerial imagery.

Within hours, users had conjured a nuclear plant in Iran, a bomb crater beside a Gaza hospital, and floods lapping the U.S. Capitol. The feature was pulled within a day following outcry from investigative experts and media organizations, with Google saying it would return with "stronger guardrails". However, the damage to trust was immediate and lasting. As one expert put it, "No benefit of integrating generative AI into Google Earth is worth the harm and the doubt it creates".

This incident reflects a broader pattern: new ways to fabricate launch overnight, a storm of criticism ensues, and guardrails arrive only after the harm. OpenAI shut down its Sora video app in March 2026 following sustained public demands for its withdrawal, but only months after a flood of abusive content involving women and problematic recreations of historical figures had already circulated.

How Are Authoritarian Regimes Weaponizing AI Video?

Beyond military conflicts, authoritarian governments are using AI video generation to fabricate political legitimacy. In Burkina Faso, where Captain Ibrahim Traoré seized power in a 2022 coup, videos on TikTok and rival platforms show him opening factories, discovering gold deposits, meeting world leaders, and foiling Western plots. These clips have racked up millions of views and raised his regional profile, but according to detailed analysis by the West African fact-checking organization Dubawa, they are completely fabricated. Videos depicting Traoré and Russian President Vladimir Putin discovering billions of barrels of oil in Burkina Faso never happened, nor did footage showing Traoré starting an aircraft manufacturing company.

The scale of AI-enabled information warfare extends beyond video to integrated systems of control. China has built what the Australian Strategic Policy Institute documented in December 2025 as an integrated apparatus of doctrines, regulations, censorship infrastructure, and AI-augmented criminal justice tools targeting ethnic minorities. A person identified through pervasive surveillance may be tried in an AI-enabled courtroom where a human judge simply reviews a machine-issued verdict, then end up in a "smart" prison that conducts "behavioral and emotional" analysis of inmates at all times. The threat, as analysts put it, "lies less in overt propaganda than in quiet erasure, when the machine that describes reality begins deciding which parts of reality may be seen".

The convergence of these trends reveals a critical vulnerability in the information ecosystem. As AI video generation tools become more accessible and realistic, the ability to distinguish authentic documentation from fabrication deteriorates. Regulatory frameworks have not kept pace; the EU AI Act, enacted in 2024, only began implementing some transparency obligations in August 2026, and implementation remains shaky with precise rules remaining voluntary rather than enforceable. Meanwhile, the U.S. government has ordered satellite companies to restrict journalists' access to imagery in conflict regions, citing wartime security, narrowing the supply of real information even as counterfeits spread.

The challenge ahead is not merely technical but institutional. Experts emphasize that the burden of proof must rest with companies deploying these systems, not with the people documenting their own catastrophes. Until AI video generation tools are built with verification, transparency, and accountability as foundational features rather than afterthoughts, they will continue to serve as weapons in information warfare, corroding the shared reality that democracies depend on.