Listen to this Post
Introduction: A New Digital Warfront
The cyber battlefield has evolved beyond hacking and surveillance — today, psychological operations are being supercharged by artificial intelligence and social media manipulation. A newly uncovered Iranian-backed disinformation campaign has begun targeting Israeli citizens, not through traditional military channels, but by attacking the public’s morale online. Spearheaded by a coordinated network centered on Telegram, this operation utilizes AI-powered tools and fake personas to spread demoralizing narratives across platforms like X (formerly Twitter). The scale and precision of this campaign reveal just how advanced state-sponsored influence operations have become, raising urgent concerns about the vulnerabilities of democratic societies in the digital age.
Coordinated Psychological Warfare: The New Arsenal of Influence
A detailed investigation by the Foundation for Defense of Democracies (FDD) has unveiled a large-scale influence operation orchestrated by Iranian actors aiming to destabilize Israeli society through social media. At the heart of this campaign is a Persian-language Telegram channel called “CAR ONLINE” with over 400,000 subscribers. The group uses this platform to instruct followers on creating fake Israeli identities on X, guiding them in AI-powered translation of Persian content into fluent Hebrew, crafting realistic usernames, and using demoralizing language to erode public confidence.
CAR ONLINE
The campaign is believed to involve up to 60,000 participants, although this figure remains unverified. Several clues tie CAR ONLINE to Iran: links to EITAA (an Iranian Telegram alternative), websites hosted on Iranian servers, and social media accounts tied to Iranian phone numbers. FDD researchers highlight that this campaign shares traits with earlier Iranian influence strategies used during the 2024 U.S. elections, indicating a repeatable, modular method of digital warfare.
What makes this campaign especially concerning is the strategic use of generative AI tools like ChatGPT to produce realistic, emotionally resonant content. Unlike spammy or bot-like behavior, these accounts exhibit enough nuance and linguistic fluency to fool both users and traditional content moderation systems. As a result, platforms like X and Meta are struggling to keep up. AI companies such as OpenAI are now being drawn into the fight, pressured to improve safeguards against abuse of their models for political manipulation.
FDD emphasizes that the answer is not just more moderation algorithms. What’s needed are deep, cross-platform investigations and international cooperation to track and dismantle these malign networks. Without a robust response, these operations could shift public opinion, erode democratic trust, and fuel division in already polarized societies.
What Undercode Say:
Digital Guerrilla Warfare Is Real and Growing
This campaign marks a pivotal evolution in the art of psychological warfare. Unlike traditional propaganda, the Iranian-backed CAR ONLINE campaign uses AI to simulate authenticity, strategically embedding false sentiment within the social media ecosystem. The goal isn’t simply to lie — it’s to make people believe they’re isolated, hopeless, and losing, by mimicking real human despair. That kind of narrative warfare strikes at the core of a population’s resilience.
Telegram: The Brain of the Operation
What’s most notable is the organizational core — a Telegram channel that not only serves as a content factory but also functions like a digital training camp. From account creation to psychological messaging and evasion techniques, CAR ONLINE provides a playbook that even inexperienced users can follow. This transforms ordinary Telegram users into foot soldiers of a larger influence war.
AI as a Weaponized Tool
Using AI tools like ChatGPT to translate content into fluent Hebrew may sound benign on the surface. But in this context, it serves a strategic function — camouflaging malicious content as local opinion. AI-generated text, especially when paraphrased skillfully, bypasses traditional moderation filters and creates a flood of indistinguishable content. This not only makes detection harder but also gives disinformation campaigns longer shelf life online.
The Power of Emotion
The messages promoted aren’t political manifestos or overt lies. Instead, they traffic in feelings — despair, hopelessness, anxiety. This is what makes them dangerous. People relate to emotional content more readily than fact-based statements, and when it appears to come from fellow citizens, it builds false emotional consensus. That consensus can tilt public sentiment and affect decision-making at a mass scale.
Strategic Duplication of Past Tactics
Iran appears to be replicating successful techniques from earlier operations, such as those targeting the U.S. during its 2024 elections. By decentralizing content creation and crowdsourcing disinformation via platforms like Telegram, these actors spread risk while maximizing reach. This modular method means the same strategy can be adjusted for different audiences — Americans, Israelis, or any other target.
The Fragility of Detection Systems
The campaign also reveals how brittle current moderation systems are. Platforms like X and Meta rely heavily on automated systems that flag repetitive, spam-like behavior. But when adversaries vary syntax, personalize content, and distribute control across thousands of accounts, these systems fail. Only deep human-led investigations — network analysis, behavioral mapping, attribution tracing — can reveal the full architecture behind such campaigns.
Implications for Global Democracies
The broader implication is that any democratic society with open digital infrastructure is vulnerable. These influence operations don’t require sophisticated cyber skills. Just a motivated state actor, AI tools, and a receptive user base. That lowers the entry barrier dramatically, enabling even mid-level actors to launch campaigns capable of swaying opinion and creating internal discord.
The Case for Public Awareness
Combatting such operations isn’t just a tech challenge — it’s also social. Users must become savvier at spotting manipulation. Education campaigns that expose tactics and show real examples could build digital immunity. Otherwise, foreign actors will continue exploiting emotional vulnerabilities and the illusion of peer opinion to divide societies from within.
Regulation, AI Governance, and Global Cooperation
The urgency is clear. AI governance must evolve quickly to address misuse. Tech companies, governments, and international watchdogs need to coordinate on threat identification, content attribution, and rapid takedown processes. Without a unified front, these malign actors will continue operating in the shadows, capitalizing on delay and disunity.
Long-Term Ramifications
If left unchecked, this pattern could evolve into something far more dangerous. AI-generated disinformation may eventually target electoral processes, military decision-making, or emergency response systems. This isn’t just a social media issue anymore — it’s a national security threat.
🔍 Fact Checker Results:
✅ The Telegram channel “CAR ONLINE” is real and publicly accessible
✅ Technical ties to Iranian infrastructure have been verified by researchers
❌ Claims of 60,000 participants remain unverified and likely inflated
📊 Prediction:
This is unlikely to be an isolated operation. In the next 12 to 18 months, similar tactics may emerge targeting elections in Europe, North America, and Asia. AI-generated psychological warfare is set to become a standard tool in state-sponsored influence campaigns — not just by Iran, but by other authoritarian regimes eager to destabilize democratic societies. The line between truth and manipulation will continue to blur, demanding a new level of digital literacy and geopolitical vigilance.
References:
Reported By: cyberpress.org
Extra Source Hub:
https://www.medium.com
Wikipedia
OpenAi & Undercode AI
Image Source:
Unsplash
Undercode AI DI v2
