Wrote: Festim Rizanaj & Xhemajl Avdyli
In the current media environment, the line between verified news and speculative content is increasingly blurred. The publication of unconfirmed materials poses a serious risk to the integrity of public information and to citizens' trust in the media.
This problem has been exacerbated by the development of artificial intelligence (AI), which enables the generation, modification or manipulation of texts, photos and videos in a way that is increasingly difficult to distinguish from authentic content. In the absence of clear editorial standards and effective control and verification mechanisms, such content, which circulates on social media, can easily become “news”, producing disinformation, confusion and, in certain cases, public panic.
A concrete illustration of this issue is the case of reports about the so-called “ghost of Prishtina”. The narrative was built on several videos initially published on TikTok, which were then reproduced by several online media without undergoing a proper verification process. The reports claimed a masked person moving around different neighborhoods of the capital, raising concerns about public safety. However, the analysis of this case highlights several serious shortcomings:
- the videos were unclear, of low quality and without elements that allowed for verification of the person's identity; in this context, it remained unclear whether the content was authentic or manipulated, including the possibility of AI assistance;
- allegations of complaints or reports from citizens were not confirmed by competent institutions, despite being presented as facts in the media;
- The reproduction of the same videos by several media outlets contributed to the viral spread of a narrative without factual basis.
AI-generated content (photos and videos) was also published and distributed during the election campaign for the December 28 snap elections, which contained unfounded political narratives. This content was amplified on social networks and widely consumed by users.
Such cases highlight the need for media regulators and newsrooms to develop clear guidelines for the use of artificial intelligence and digital content. These standards should include mandatory verification of the origin of visual materials, transparency regarding technological interventions, and adherence to the fundamental principles of responsible journalism. In the absence of a clear regulatory framework, AI-generated or manipulated images and videos risk being used to construct fictional narratives that appear convincing.
In terms of legal and regulatory aspects, Kosovo currently does not have a specific framework that directly addresses the use of AI in the media or the obligations for verification of content generated or manipulated through it. The existing regulatory framework for media and audiovisual communication relies mainly on general laws on media, broadcasters and electronic communications, as well as on ethical standards of self-regulation, which do not explicitly address synthetic content and new digital challenges. Meanwhile, although Kosovo has expressed an orientation towards approximation with European Union legislation, these frameworks have not yet been transposed into domestic legislation and by-laws. This creates a regulatory gap, where the responsibility for verification and transparency falls mainly on individual newsrooms.
Maintaining the integrity of public information requires that facts and verification prevail over virality, ensuring that technology serves as a tool for quality information and not as a source of uncertainty and misinformation.