The rise of artificial intelligence has ushered in a disturbing new era for the media industry, as bad actors increasingly leverage the technology to perpetrate violence and harassment against journalists. According to data from Free Press Unlimited and the Mapping Media Freedom (MMF) Monitor, there were 58 documented incidents between 2022 and 2026 where AI was explicitly used to obstruct or intimidate members of the press across Europe. This “AI-facilitated violence” marks a significant escalation in how digital tools are being weaponized to erode the credibility of reporters and destabilize the public’s trust in journalism.
A primary weapon in this arsenal is the AI-facilitated scam, which exploits the hard-earned professional reputation of journalists to defraud the public. Seventeen of the 58 recorded incidents involved deepfake advertisements where journalists’ likenesses were used to promote fraudulent investment schemes. A harrowing example involves Swedish financial journalist Gabriel Mellqvist, whose image was hijacked in a deepfake ad, resulting in roughly 500 investors losing approximately 46 million euros. These scams not only inflict financial ruin on victims but also inflict long-term damage on the credibility of the targeted journalists, whose names become synonymous with the fraud in the eyes of the public.
Beyond individual financial crimes, AI is serving as a critical infrastructure for state-sponsored disinformation campaigns, most notably those orchestrated by Russian interests. Campaigns like “Portal Kombat” and “Doppelganger” have utilized AI to mass-produce deepfake videos and clone entire news websites to spread pro-Russian narratives. One high-profile case involved French news anchor Julien Fanciulli, whose likeness was used in a deepfake to fabricate a report about an assassination attempt on President Macron. These networks, which researchers estimate have disseminated over 8 million fake articles since 2023, effectively use AI to perform “information laundering,” making it nearly impossible for the average reader to distinguish official reporting from state-backed propaganda.
Perhaps the most dehumanizing aspect of this trend is the surge in sexualized AI-generated content targeting female journalists. Five of the 58 analyzed cases involved non-consensual sexual deepfakes, a phenomenon that aligns with broader global trends where an estimated 96–98% of all deepfakes are pornographic in nature, with women accounting for 99% of victims. Experts note that these digital attacks are not mere nuisances; they mirror the psychological trauma of physical sexual violence. Because these images are persistent and difficult to purge from the internet, they serve as a weapon of attrition, forcing many women to withdraw from public life and the journalistic profession entirely.
Political intimidation also utilizes AI to incite real-world harm. In Serbia, journalist Dinko Gruhonjić became the target of a sophisticated video manipulation that falsely portrayed him as a supporter of war criminals. The video, which was amplified by both nationalist actors and mainstream media outlets, triggered a wave of death threats and physical safety concerns for the reporter. This incident highlights a dangerous feedback loop where AI-generated content is weaponized by political extremists to incite hostility, effectively creating a “digital bounty” on the heads of journalists who are already at risk due to their critical reporting.
In response to this evolving threat landscape, Free Press Unlimited emphasizes the urgent need for both institutional policy changes and individual vigilance. The organization advocates for a three-pronged approach: fostering critical media literacy among the public to identify potential manipulations, utilizing advanced forensic tools to verify content, and maintaining a consistent record of reporting all incidents to monitoring bodies like the MFRR. As AI-generated content becomes increasingly indistinguishable from reality, the protection of press freedom will require not just new legislation, but a collective societal refusal to be manipulated by the digital shadows cast by these technologies.