Reporting harmful content

Recognising the scale of harmful content circulating on major social media platforms in conflict contexts — and the risks this poses not only locally but across European information spaces — IMS has been engaging with Appeals Centre Europe (ACE) as part of its broader commitment to formal, rights-based redress mechanisms. This engagement reflects IMS’s intent to use the tools available under the EU’s Digital Services Act (DSA) framework to surface content moderation issues that carry systemic implications for media integrity, civic discourse and public safety.

Through this process, IMS has identified several recurring categories of harmful content, originating in or directly linked to the context of Russia’s full-scale war against Ukraine.

While these risks are acutely felt within Ukraine, the content in question circulates freely across borders — accessible to audiences throughout Europe and beyond — and therefore carries implications that are inherently transnational in nature.

The first category concerns synthetic and AI-generated disinformation: fabricated content designed to manipulate public perception of the conflict, distort factual narratives, and undermine trust in legitimate institutions and actors. The second covers incitement to hatred and glorification of violence: content that frames military aggression in celebratory or normalising terms, including material with symbolic references to documented atrocities.

The third concerns coordinated inauthentic narratives targeting public figures, state symbols and minority communities — including content that demeans or threatens individuals on the basis of gender identity and sexual orientation. The fourth relates to content that, while not always meeting the threshold of formal policy violations under current platform definitions, exhibits clear indicators of organised, cross-border information operations designed to erode social cohesion and public trust.

These patterns have been observed repeatedly across multiple appeal cases reviewed through this engagement, indicating that the issues described are not isolated incidents but reflect recurring dynamics in platform governance responses to conflict-related content.

These categories correspond directly to the systemic risk typologies under Article 34 of the DSA — encompassing threats to public security, fundamental rights, civic discourse, and the dissemination of illegal content including hate speech and incitement to violence. The transnational reach of this content makes it a concern not only for Ukraine but for the integrity of information environments across EU member states.

What this experience has also revealed is a consistent and notable gap between formal dispute outcomes and actual platform behaviour. In IMS’s assessment, this gap is indicative of structural limitations in current moderation and enforcement frameworks, including insufficient conflict-sensitive risk analysis and delays in translating appeal outcomes into timely platform action.

In a significant share of cases, content remained available even following decisions issued in favour of the appellant. This pattern — low implementation rates combined with platform policies that do not adequately account for the specific dynamics of information warfare — points to a structural challenge that extends well beyond individual content decisions and warrants attention at the systemic and regulatory level.

There are nuances as certain content categories, such as straightforward policy violations in Moldova receive a better platform response than others such as disinformation in war contexts like Ukraine, where platform response remains very limited.

These observations may therefore be relevant for the consideration of systemic risk assessments, mitigation measures, and implementation monitoring under the DSA, beyond the resolution of individual content-level disputes.

Download the full report here: Transparency-Report-May-2026-English.pdf

CASE

Manipulated News Story About Ukraine Russia War

In March 2026 IMS submitted a dispute about a video on TikTok for violating its misinformation policy. The video, which is visible in the EU, shows a newsreader talking about Russia launching missiles at Ukraine. The video has clearly been manipulated with AI. The voiceover has been added by the person who posted the content and the words have been artificially synchronised to the presenter’s speech. The artificial voiceover and editing meant that the video was using a trusted news source to mislead people on a matter of public importance. For this reason, we decided it violated TikTok’s policy on AI-generated content and overturned the platform’s decision to leave it up. We then sent our decision to IMS and TikTok. As of mid-April, the video remains online and TikTok has not implemented our decision. IMS has observed a significant number of similar AI-generated videos being posted recently and believes this represents a broader pattern of AI-manipulated content related to the war between Russia and Ukraine.