Democracy Reporting International

Report September 2026

Submitted

Democracy Reporting International (DRI) is an independent organisation dedicated to promoting democracy worldwide. We strengthen democracy by supporting the institutions and processes that make it sustainable and work with all stakeholders towards ensuring that citizens play a role in shaping their environment. DRI focuses on five key themes on democracy: justice, elections, local governance, digital democracy and human rights. DRI is financed by and partners with a diverse group of foundations, firms and governments, including: the German Federal Foreign Office, The Ministry of Foreign Affairs of the Netherlands, the European Union, Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), the Mercator Foundation, the Landecker Foundation and the Porticus Foundation. Our team of researchers, project managers and policy experts bring diverse backgrounds in quantitative data analysis, advocacy, law, policy and media. In addition to the colleagues listed below, our team is currently recruiting new policy and outreach officers to support our endeavours. Our global network of experienced democracy champions, both within and outside the organisation, refine and compliment this expertise to allow for multistakeholder research and analysis into online democratic discourse. In order to combat disinformation, Democracy Reporting International's Digital Democracy (DD) experts conduct research into the tools, tactics, and technologies used to disseminate disinformation. Furthermore, they develop material that assists policymakers, researchers, and practitioners in detecting even advanced forms of disinformation, such as those produced by generative artificial intelligence. With social media monitoring, especially in the context of elections, being a core competency of the DD team, DD experts identify patterns of information manipulation. In addition, they provide training for other stakeholders to conduct extensive social media monitoring.

Crisis and Elections Response

Elections 2025

[Note: Signatories are requested to provide information relevant to their particular response to the threats and challenges they observed on their service(s). They ensure that the information below provides an accurate and complete report of their relevant actions. As operational responses to crisis/election situations can vary from service to service, an absence of information should not be considered a priori a shortfall in the way a particular service has responded. Impact metrics are accurate to the best of signatories’ abilities to measure them].

Threats observed or anticipated

1. AI-generated content in electoral contexts: policy and enforcement gaps 
During the reporting period, DRI assessed how VLOPs govern AI-generated content in political contexts, as generative AI reshapes political communication and lets political actors create and spread synthetic content at scale and speed. VLOPs increasingly distribute this content, yet their policies remain inconsistent, opaque, and weakly enforced. We examined Meta, TikTok, and YouTube across five dimensions: actors, behaviours, content, degree, and effect. None of the three platforms mandate official political accounts to label AI-generated content, even though parties, candidates, and government actors reach larger audiences and carry greater authority than ordinary users. Most platforms do not treat AI-generated manipulative political content as a distinct electoral risk. TikTok addresses it under its integrity guidelines, but Meta and YouTube fold it into broader manipulated-media and misinformation frameworks. Automatic detection and labelling are applied inconsistently, with no clear triggers, leaving significant gaps in coverage. 

We tested these gaps empirically during Hungary's 2026 elections. Using TikTok's search function and election-related hashtags, we identified 2,043 posts from 22 accounts that reached over 50 million views and either posted predominantly AI-generated content or presented as AI-generated personas, focused on Hungarian politics. Nearly three quarters of the posts (71.6 per cent) contained AI-generated content, and 37.5 per cent of that content was left unlabelled by the user or the platform. We reported 199 unlabelled posts to TikTok for non-compliance with its Edited Media and AI-Generated Content labelling requirements. The company removed 19 videos and one account and restricted the visibility of two posts. The remaining 177 posts stayed unlabelled despite our flagging.  

Additionally, we submitted comments to Meta's Oversight Board on generative AI use during the elections. We argued that Meta's labelling standard sets a higher bar than the DSA Electoral Integrity Guidelines and called on Meta to align with the guidelines, cover the full electoral cycle, and sanction non-disclosure. 

 
2. Cross-border amplification and inauthentic networks on X 

DRI investigated inauthentic behaviour on X in the context of Hungary's 2026 parliamentary elections. This research was made possible by a Berlin Court of Appeal ruling in February 2026, which required X to grant DRI data access through its API. Querying data at scale is what allowed us to detect patterns of coordination that manual browsing could not surface. Our initial observations identified 46 suspicious cryptocurrency and NFT-themed accounts commenting on Hungarian election posts. These accounts otherwise posted in English about crypto but switched to Hungarian to criticise opposition leader Péter Magyar and support Prime Minister Orbán and his Fidesz party. They amplified two posts attacking Magyar, one of which traced to the Russian Storm-1516 disinformation campaign. X's moderation response was inconsistent: when we reported 13 of the accounts, the platform stated they were not in violation of its rules, yet 45 of the 46 accounts were later suspended. 

We subsequently identified a larger, higher-confidence network of 337 accounts based in Serbia. All were created in 2026 and amplified content promoting Orbán and Fidesz while spreading disinformation about Magyar, whose Tisza party ultimately won the election. The network showed clear automation signals: bulk account registration, mathematically regular posting intervals, duplicate messaging, and concentrated burst activity between 9 and 11 April. In several cases, these accounts made up over 30 per cent of a post's reposts, quote tweets, or replies. The network appeared primarily focused on Serbian domestic politics and switched to Hungary only temporarily, pointing to cross-border amplification. 

 
3. Cross platform Information dissemination and Inauthentic Networks on Telegram 
During the reporting period, DRI developed a replicable methodology to measure how coordinated information manipulation moves across platforms. FIMI campaigns increasingly operate across several services at once, exploiting regulatory gaps and evading detection. Messaging apps sit at the centre of this, with Telegram often serving as the coordination hub where narratives form before spreading to mainstream platforms. We piloted the methodology during the Dutch snap parliamentary elections of October 2025, tracing how Telegram content surfaced on Facebook and TikTok. 

We applied this Telegram focus directly to Hungary's 2026 elections, where we uncovered a coordinated network of 19 inauthentic news channels operating across many languages, including most official EU languages, with a combined subscriber base of 167,000. The channels shared clear signatures of automation and coordination. Most relied on the same contact bot, posted at fixed intervals, and published within the same minute of one another. 

Mitigations in place

1. Code of Conduct on Disinformation 
DRI continued its reporting efforts under the Rapid Response System of the Code of Conduct on Disinformation, actively collaborating with signatories. Between January and February 2026, DRI helped shape new transparency guidelines for the Rapid Response System under the Code. Co-chairing the Working Group on Elections alongside TikTok, we led negotiations for the non-platform signatories, producing a public Q&A on how the system works and internal guidelines on how signatories may display the content they report.

2. Raised awareness about threats and built networks with relevant stakeholders 
On 16 July 2026, DRI hosted an online roundtable on countering FIMI and accountability gaps regarding Telegram. The event brought together more than 35 participants, including researchers, regulators, and fact-checking experts, to exchange approaches to detection and attribution and to examine regulatory gaps and enforcement challenges. The first session focused on detecting and attributing coordinated inauthentic behaviour on Telegram, drawing on collaboration with CeMAS (Center für Monitoring, Analyse und Strategie) and GLOBSEC. Discussion centred on how researchers surface coordination through signals such as matching channel names, synchronised posting, and improbable channel growth, and on the constraint that platform-level data access continues to place on this work. The second session brought together the European Fact-Checking Standards Network (EFCSN) and the Belgian and Slovak Digital Services Coordinators to examine accountability and the regulatory gaps around Telegram and other non-VLOP platforms. A recurring theme was that Telegram's lack of VLOP designation lets it escape the systemic risk obligations that would otherwise apply under the DSA. 

We developed this argument further in an op-ed, drawing on our research into a multilingual Telegram network spreading Kremlin-aligned disinformation. We argued that Telegram's reach in the EU warrants closer scrutiny of its reported user numbers, and that VLOP designation would require it to assess and mitigate the systemic risks tied to coordinated manipulation. 


 3. Communicating continued data access challenges 
DRI continued to raise awareness about barriers to platform data access under the DSA. On 25 June 2026, we participated in a webinar hosted by EU Disinfolab on our litigation against X to obtain platform data under Article 40(12) of the DSA. We shared how the legal process unfolded and what the ruling means for platform accountability and independent research. We also offered practical lessons for researchers and civil society actors considering similar action, at a moment when the European Commission had itself found X non-compliant on related grounds.