Russia's AI videos surge as Western defenses lag behind

Security experts warn that Moscow-aligned actors are exploiting recent advances in generative video to flood social platforms with hyperreal, emotionally charged clips aimed at shifting public opinion across Europe and the United States, and that Western governments are poorly equipped to stop it.

The technology at the center of the campaign can create believable footage of public figures, stitch together fabricated events and repurpose real scenes with altered audio in minutes. That speed and scale collapse the traditional cost and time barriers that once limited deepfake production to black-ops labs and cinematic houses. Analysts say this has turned synthetic video into an accelerant for older disinformation playbooks: narrative saturation, identity-based targeting and rapid re-posting through networks of domestic proxies and foreign media.

Platforms are struggling to keep pace. Content moderation teams rely on a mix of automated filters and human review, but generative video upends both approaches. Automated detectors trained on older manipulations generate false negatives against novel model outputs, while human reviewers cannot feasibly check the millions of clips uploaded daily. Cross-border distribution further complicates takedowns: a clip seeded in one country can be mirrored, translated and amplified in many languages before any single regulator or platform can act.

Western legal frameworks provide limited help. The European Union's Digital Services Act imposes transparency and risk-mitigation obligations on major platforms, but enforcement mechanisms are still developing and do not directly criminalize state-directed synthetic-messaging campaigns. In the United States, federal law contains few provisions tailored to state-backed information operations, leaving responsibility diffuse across agencies and often dependent on platform cooperation. International law remains vague on the sovereignty and attribution issues created by synthetic media used to manipulate foreign electorates or incite unrest.

The geopolitical stakes are immediate. Security officials caution that synthetic video can harden social divisions, discredit democratic institutions and complicate crisis escalation management by producing plausible but false evidence of events on the ground. For societies with polarized media ecosystems or large diasporas tied to homeland politics, the tactic is particularly potent: tailored content in native languages reaches receptive audiences with high trust thresholds, making correction and context interventions slow and ineffective.

Experts propose a clustered response that combines technology, policy and public diplomacy. Technically, governments and industry should invest in shared forensic tools and open threat exchanges to speed attribution and detection. Policy measures could include mandatory provenance labeling for synthetic media, accelerated transparency requirements for political advertising and clearer sanctions mechanisms for state actors that deploy disinformation as an instrument of foreign policy. Diplomatically, coalitions of like-minded states could coordinate rapid rebuttals and legal actions, while funding independent local journalism and media literacy programs would blunt impact at the societal level.

Even with those steps, observers say the balance of advantage now favors the sender. Generative video lowers barriers to entry for state and nonstate manipulators alike, and the velocity of information ecosystems makes retroactive correction difficult. As the technology spreads beyond a handful of competent operators, democracies will face a prolonged contest over credibility, one that will test not just cybersecurity and law, but cultural resilience and the transatlantic will to cooperate.