Fundación Maldita.es has contributed to the second annual report on the most significant and recurring systemic risks affecting very large online platforms and search engines.
The European Board for Digital Services, composed of the digital regulators of each EU Member State, is responsible for producing this annual report in collaboration with the European Commission under the Digital Services Act (DSA).
The report examines both the main systemic risks associated with the design and operation of these online services and the best practices for reducing them. This approach reflects the obligations set out in the DSA, which requires very large online platforms and search engines to identify these risks and implement mitigation measures that are reasonable, proportionate, effective, and tailored to the specific characteristics of each risk.
Following the publication of the first report of this kind in November 2025, Fundación Maldita.es has once again contributed evidence for the preparation of the second report, which covers the period from February 2025 to February 2026.
Generative AI and economic incentives
In our contribution to the report, we focused on the impact of AI-generated content. These tools facilitate the creation and dissemination of coordinated disinformation campaigns, make false content appear more credible, and create new risks for children's safety, for example through the creation of videos that sexualize them. We also warned that we continue to find illegal content on platforms such as Facebook and TikTok.
We also identified several factors that contribute to these risks. One of them is the financial incentives offered by some platforms through creator programmes or subscription systems, which may end up rewarding the publication of harmful content. We also pointed out that recommendation systems contribute to amplifying disinformation, as seen during the DANA floods, as well as videos that sexualize children. This is compounded by the role of advertising systems on very large online platforms and very large online search engines (VLOPs and VLOSEs), which enable this content to be targeted at specific audiences while the companies profit financially. Finally, we highlighted that inconsistencies in content moderation persist, making it easier for this type of content to remain available and continue spreading even after it has been reported.
Risk mitigation on platforms and search engines
We also stressed the importance of providing context alongside false or misleading content. Various studies show that adding verified information alongside this type of content helps reduce its spread. For this reason, we examined both collaboration programmes with fact-checking organisations and the advantages and limitations of Community Notes, a tool that an increasing number of platforms are adopting.
Although compliance with the DSA is the responsibility of each platform individually, many of the risks we identified are shared. Disinformation campaigns often operate across several platforms at the same time, meaning that the lack of mitigation measures on one platform ultimately affects the others. One example is the limited interoperability between platforms. AI-generated content that is labelled on one platform can be downloaded and reposted on another, where that warning disappears. In addition, there are also risks stemming from platforms that have not been designated as VLOPs, but whose lack of mitigation ultimately has an impact on the rest of the online ecosystem.
Full report: