Russia and China have coordinated AI-driven influence operations targeting European countries to erode support for Ukraine, generating hundreds of millions of views with false narratives about the conflict bankrupting Europe. However, a major setback in Hungary's April 2026 elections reveals that robust democratic engagement and independent media can overcome even sophisticated propaganda machines powered by artificial intelligence.

What Are Russia and China Actually Doing With AI in Europe?

The shift began in 2025 when U.S. military aid to Ukraine dropped sharply and European nations stepped in to fill the gap. Russia's influence operation network, Storm-1516, which is linked to the Internet Research Agency, pivoted its focus from undermining international support for Ukraine to targeting the European countries now bankrolling the conflict.

Germany and France became primary targets. Germany emerged as Ukraine's largest military donor at €9 billion, while France's president positioned himself as one of Kyiv's leading champions. According to a February 2026 investigation by NewsGuard, Storm-1516 published 34 false claims aimed at these two countries between January 2025 and February 2026, generating 274 million views. The core message remained consistent: Ukraine is bankrupting Europe, exhausting its citizens, and dragging the continent toward war.

What makes this campaign particularly concerning is the deepening coordination between Moscow and Beijing. The two countries have coordinated on artificial intelligence governance, digital standards, and media policy, and this partnership now extends to influence operations. OpenAI's October 2025 threat report found Russian-linked actors using the Chinese AI model DeepSeek to conduct influence campaigns. Independent analysis also showed that Chinese AI models embed content controls that reproduce Kremlin narratives on Ukraine. The European Commission's 2025 report on foreign information manipulation concluded that Russian and Chinese actors have "fully implemented AI tools" for such operations.

Where Are the Gaps in Europe's AI Regulation?

Europe's regulatory framework, including the EU AI Act and the Digital Services Act (DSA), is designed to combat such operations. However, significant gaps allow propaganda to circulate freely. The DSA designates X as a very large online platform and requires it to identify and block disinformation. European regulators fined the platform €120 million in December 2025 for breaching DSA transparency obligations, and the Commission extended its investigation into X's recommender systems in January 2026.

But the same narratives migrate within hours to platforms that fall outside regulatory reach:

• Telegram's Exemption: The messaging service is not designated as a very large online platform, has not signed the EU's Code of Practice on Disinformation, and offers no system for reporting disinformation. Ukraine's Centre for Democracy and Rule of Law found that sanctioned Russian propaganda channels on Telegram are blocked only partially, while duplicate channels carrying identical content operate freely under altered names.
• Transatlantic Regulatory Gap: Europe's DSA applies to platforms operating in the EU, but U.S.-hosted services face no equivalent obligations. The current U.S. administration has widened this gap by rolling back AI safety commitments and framing European digital regulation as protectionist. A Ukraine fatigue narrative manufactured in Moscow can move through U.S.-based channels and resurface in European media looking like organic public opinion.
• Balkan Border Gap: Candidate countries in the Western Balkans have no obligation to implement the DSA and have limited capacity to monitor AI-generated content. The same messaging telling French and German voters their governments are wasting money on Kyiv gets repackaged for Balkan audiences as evidence that the EU cannot sustain its commitments, undermining trust in EU enlargement promises.

These three gaps converged during Hungary's April 2026 elections. The ruling Fidesz party ran a campaign built on Ukraine fatigue, reinforced by AI-generated imagery, deploying synthetic news anchors and deepfake endorsements on TikTok and content on X and Telegram.

How Did Hungary's Voters Defeat the Propaganda Machine?

Despite the coordinated foreign campaigns, the AI-generated content, and the bot networks, the propaganda failed spectacularly. Hungarian voters saw through the lies and rejected Viktor Orbán in a historic landslide. Péter Magyar's Tisza Party won 141 out of 199 seats, ending 16 years of Fidesz rule.

The lesson from Hungary challenges a common assumption among European policymakers: the most effective counter to AI-powered propaganda is not better regulation alone, but better democratic speech. Credible counternarratives from independent media, fact-checkers, and political leaders willing to make the case for Ukraine in plain language did more to defeat Fidesz's narrative machine than any platform rule.

Steps to Strengthen Defenses Against AI-Driven Disinformation

• Invest in Independent Journalism: Scaling credible counternarratives from independent media outlets ensures that truthful voices are louder and faster than fabricated ones, providing citizens with reliable information sources.
• Expand Civic Media Literacy: Teaching citizens to recognize AI-generated content, deepfakes, and coordinated disinformation campaigns helps voters evaluate claims critically and resist manipulation.
• Enforce Existing Regulations: Bringing Telegram under meaningful DSA obligations, enforcing the AI Act's transparency rules, and helping Western Balkan candidate countries build fact-checking capacity can narrow the gaps that adversaries exploit.

Regulation still has a role, but a more modest one than Europe's policymakers typically acknowledge. The DSA's record of only two fines in over two years is a reminder that rules on paper mean little without the will to apply them. What defeated the propaganda in Hungary was not an algorithm or a regulation. It was citizens who refused to believe the lies.

What Does This Mean for the EU AI Act?

The EU AI Act came into full effect on August 2, 2026, representing the most comprehensive AI regulation anywhere in the world. The law imposes fines of up to €35 million or 7 percent of global revenue for non-compliance. Its reach is extraterritorial, applying to any organization whose AI systems are utilized in the EU or whose outputs affect EU citizens and organizations.

The regulation establishes a risk-based, tiered system where the highest level of regulation applies to systems that pose the greatest risk to health, safety, and human rights. The highest risk category is now defined by the number of computer transactions needed to train a machine, known as floating point operations per second. Lower tiers place major new obligations on AI services, including strict disclosure rules about data used to train systems, restrictions on facial recognition technology except for narrowly defined law enforcement exceptions, and bans on AI for "social scoring" and systems that "manipulate human behavior to circumvent their free will".

About 47 percent of organizations citing the European regulation in governance disclosures are not headquartered in the EU, though many have significant EU market presence. In the tech sector, 53 percent of U.S. companies cite the EU legislation, reflecting the fact that leading players such as Google, Microsoft, xAI, and OpenAI have voluntarily adopted key parts of the EU's AI framework to maintain access to the 27-member state market.

However, the regulation's strict approach raises questions about Europe's broader competitiveness. Former European Central Bank chief Mario Draghi's competitiveness report highlighted how the EU's ambition to strengthen its role as a tech leader is being undermined by increased regulation. While the AI Act reinforces the "Brussels effect" of European legislation becoming a de facto global standard, early evidence indicates it may be doing little to help the bloc fulfill wider goals of boosting tech-led innovation and productivity.