Addressing the impact of artificial intelligence, cyberviolence, and deepfakes on equality, democracy and inclusive societies

The rapid development and deployment of AI technologies, along with their malicious use such as the creation of artificially generated and manipulated content,, profoundly impacts democracy, equality, social inclusion, and knowledge production, and contributes to the rise of cyberviolence. Although AI can offer many opportunities, its misuse can facilitate/amplify inequalities and power imbalances, spread misinformation and cyberviolence against women, children and minorities, concentrate data control, and undermine democratic processes. Hence, this topic addresses both the malicious use of AI (e.g. AI-generated disinformation, deepfakes, cyberviolence) and its unintentional societal consequences (e.g. algorithmic opacity, amplification of bias).

AI’s role in media influence and public trust demands urgent attention, as the rise of AI-generated or manipulated content, including deepfakes, threatens democratic processes and epistemic rights. A key area of concern is how AI and artificially generated content is impacting cyberviolence, and the effect it has on individuals or groups who distrust democracy, particularly where this distrust intersects with far-left and far-right populism and foreign interference.

Proposals should explore how AI-driven technologies and their use may either exacerbate or mitigate[[Including how trustworthy and human-centric AI can be leveraged to support content moderation, fact-checking, and online monitoring.]] inequalities and discriminations based on sex, gender, sexual orientation, racial or ethnic background, religion or belief, age, and disability, particularly in access to information, decision-making, and representation. A critical analysis is needed of how data - often reflecting existing social biases - can reinforce or challenge dominant social structures and safety, and how such technologies shape public perceptions and knowledge production.

The research should also explore the risks these technologies pose to democratic integrity, particularly through the manipulation of public opinion, electoral processes, and governance systems. Deepfakes and AI-generated misinformation and disinformation can distort facts, spread false narratives, and undermine public trust. Addressing these concerns requires a comprehensive understanding of how AI may be (mis)used to shape political campaigns, media narratives, public engagement, and the spread and dissemination of (gendered) disinformation. Proposals are encouraged to examine how AI-generated content across media, including entertainment media, influence public opinion, social narratives, gender stereotypes and norms, and civic engagement.

Furthermore, proposals should look into how cyberviolence, including online harassment, cyberbullying, threats, and gender-based violence, is exacerbated by sharing manipulated content like deepfakes, deep nudes and AI-driven sextortion on online platforms through manipulated content like deepfakes, deep nudes and AI-driven sextortion. Research should identify how AI may be used to amplify and upscale harm, what the nature of the AI-powered output is, and which groups are disproportionately affected by it, looking particularly at women and minorities, and explore interventions to prevent and mitigate these risks. This includes analysing not only the unintended reproduction of cyberviolence through biased outputs, but also the ways in which AI-powered platforms may be misused by users to generate and/or spread harmful, discriminatory or violent content. A comprehensive assessment of unregulated AI-induced risks of sexual exploitation, violence, and gender-based harm is largely missing for effective regulation, oversight, and prosecution. A multidisciplinary review is needed to evaluate AI-driven risks exposing children, young people, women, older persons and LGBTIQ people to such violence online and offline.

Proposals should research on policy and concrete practices that can effectively address the challenges posed by AI technologies, taking into consideration the EU Regulation on artificial intelligence and subsequent guidance being developed to support its implementation. Proposals should identify best practices and regulatory measures to ensure the ethical deployment of AI and AI literacy while safeguarding equality and democratic integrity. A key aspect of this research is recognising AI’s growing role in shaping policy and judicial decisions. While its integration can improve efficiencies, concerns about bias and fairness persist. Proposals are encouraged to explore AI’s influence on legal and policy outcomes, including unintended consequences.

Additionally, proposals should consider AI’s use and impact on young people. Although AI is not specifically designed for minors, youth are among its most active users. Biases within AI-generated or manipulated content, particularly related to sex gender, sexual orientation, ethnic and racial background, can shape young users’ perceptions, perpetuate stereotypes, affect engagement and mental and physical well-being. Research should investigate how AI systems influence these aspects, and the opportunities and risks associated, including the potential psychological effects of exposure to biased or harmful content.

A key focus should be interdisciplinary research on AI’s role in societal resilience, countering misinformation and disinformation, enhancing civic engagement, and supporting marginalised communities. Proposals should explore inclusive and innovative tools and methodologies for detecting and mitigating deepfakes, disinformation, and cyberviolence facilitated through AI, designed for broad adoption by policymakers, technology developers, media organisations, and the general public.

Effectively addressing these multifaceted challenges demands to combine data-driven analysis, drawing on expertise from a wide range of fields, including both academic disciplines (e.g. computer science, and SSH disciplines such as communication and media sciences, ethics, law, political science, sociology, psychology, and gender studies) and the applied perspectives of those involved in shaping and steering AI technologies in practice.

This research should contribute to the EU’s broader AI strategy, supporting the implementation of the Artificial Intelligence Act and aligning with the EU’s Political Guidelines for 2024-2029. This research should also contribute to the implementation of other EU legislative frameworks, such as the Directive on combating violence against women and domestic violence, that criminalises various cyberviolence offences, including malicious deepfakes, and the Digital Services Act, which addresses illegal content online, including false or manipulated information inciting hate or discrimination. It should also inform global discussions on ethical AI governance and responsible innovation.

Proposals are encouraged to identify other relevant EU-funded projects, and to explore potential collaboration opportunities with them.

The projects selected for funding are encouraged to collaborate with the JRC to seek synergies with its work on innovations in public governance[[ https://joint-research-centre.ec.europa.eu/projects-and-activities/innovations-public-governance_en]].