DesinFact – Early detection of disinformation

Katharina Kloiber | 05.12.2025

Get in touch

From January 2024 to December 2025, the DesinFact project explored approaches to identifying trends at an early stage and evaluating them in terms of disinformation.

Disinformation poses a significant challenge to our society. Concerted disinformation campaigns are one aspect of hybrid threats, which can target specific critical infrastructure, such as the secure distribution of energy sources, raw materials, or medicines, with the aim of disrupting or damaging it. Alternatively, they can target broader critical infrastructure, such as democratic institutions, with the aim of undermining them and destroying trust in them or their representatives.

The timely detection of disinformation campaigns therefore makes an essential contribution to resilience in the face of such threats. Currently, however, there are hardly any tools available to actively detect disinformation campaigns at an early stage. Those affected often find out about their involvement far too late, which limits their ability to respond effectively. Often, the only option left is damage control. Early detection of such trends would provide scope for action, e.g. to develop appropriate counter-statements.

DesinFact aims to improve the state of research on the automatic detection of disinformation trends, identify gaps in technical, legal and ethical areas, and develop suitable approaches to enable such a system.

Research into the provision of a public system for detecting disinformation

A public system should enable citizens to have online content checked for disinformation. In DesinFact, we have therefore set up a prototype browser plugin and dashboard that allows users to query website content and displays analysis results. In close cooperation with our project partners, we have addressed socio-technical aspects that are relevant for the adequate introduction of such technology.

It has been shown that providing a system for detecting disinformation is not a problem from a technical point of view and largely depends on the available hardware resources. Both user anonymity and search histories as well as the evaluation of trends can be implemented technically. Additional manual quality control and appropriate processes also adequately prevent misuse.

However, the situation is different when it comes to analysing the content and presenting the results. The biggest challenge is to present results as a decision-making aid without making any judgements. Since disinformation is a highly complex issue whose assessment depends on numerous factors, controversial decisions are almost inevitable. Accordingly, both the evaluation systems and the presentation of results must be clear and comprehensible, as has been confirmed in the course of the project.

Interdisciplinary collaboration

The DesinFact project is funded by the Austrian Research Promotion Agency (FFG) as part of the KIRAS programme. The following project partners are conducting joint research into ways of detecting disinformation at an early stage:

description