AI Companies and Platforms: How AI Is Changing Influence Operations

State-backed propaganda networks across Iran, Russia, and China are deploying frontier artificial intelligence to draft operational doctrines, manage overseas media assets, and automate cyberattacks, according to threat reports released by frontier AI companies and social media platforms. While the technology has sharply lowered the cost and increased the output of digital influence campaigns, platform disclosures indicate that these automated operations have largely failed to alter public opinion or achieve meaningful audience reach.

Iran Uses AI for Propaganda and Funeral Planning

Iranian state propagandists used Anthropic’s Claude AI model to draft the doctrine for their influence campaigns and to plan the funeral of the late Supreme Leader, according to industry disclosures cited by lawfaremedia.org. In another instance, operators linked to the Iranian opposition group MEK ran a shared AI agent whose memory files stored banned words, approved sources, and detection-evasion rules to maintain continuous output without manual direction.

Russia Employs AI for Content Generation and Cyber Intrusions

Russian state media outlets utilized artificial intelligence for both content generation and administrative tasks ahead of international political events, lawfaremedia.org reported. A former Sputnik Moldova editor-in-chief used AI models to churn out articles and push fabricated claims regarding Moldova’s president ahead of the country’s 2025 election, while a staffer at a Russian state media outlet employed an AI model to write on-air chyrons. Beyond influence work, Google and Anthropic reported that Russia-linked groups used AI across cyber intrusion efforts. In one case, a Russia-linked actor used AI agents to monitor whether security products had detected its malware, modifying and rebuilding the code automatically until it bypassed defenses.

Chinese Influence Operators Target American Energy Debate

China-linked operators used American artificial intelligence systems to meddle in domestic U.S. debates regarding AI and infrastructure costs. OpenAI reported in June that accounts likely run by a Chinese tech contractor falsely claimed ChatGPT had leaked user data, while another cluster generated political cartoons blaming data centers for rising American electricity bills.

Production Capacity Versus Persuasion Realities

Threat reports show that generative AI scales content while agentic AI scales operations, transforming back-office tasks like audience research, timing strategies, and workflow automation. Despite these efficiency gains, the reports indicate a persistent gap between production capacity and actual persuasion. The widest authentic reach occurred only when operators plugged AI-generated content into existing state-media distribution networks, such as a Wagner-founded FM radio station in the Central African Republic. Conversely, a French advertising network using automated tools to publish roughly 9,000 articles across 70 fake local news sites drew almost no real public engagement.

Researchers Struggle to Assess AI Influence on Public Opinion

Social media platforms and AI companies continue to track how automated influence apparatuses adapt their distribution methods by attempting to cultivate trustworthy everyday personas or by co-opting established human influencers. Assessing the downstream effect of these AI-driven operations remains a complex challenge for researchers, as high production volumes do not reliably translate into shifted public opinion or sustained audience trust.

More on this story: Israel-Backed Influence Campaign Targets Gen Z and AI Chatbots · How AI Vendors Use Security Incidents as Stealth Marketing