About the project
Disinformation and fake news have been declared one of the biggest current threats to democracy by organizations including the EU, NATO, and the UN. This threat has increased in line with the rapid development of generative artificial intelligence (AI).
There is an urgent need for:
- knowledge about what the next-generation disinformation and influence operations might look like
- a knowledge base to develop tools that can identify and warn about such disinformation
The project "NxtGenFake" will address these knowledge gaps.
Purpose
Linguists, media scholars, and computer scientists will collaborate to illuminate the following research questions:
- To what extent, and how, does AI-generated disinformation differ from disinformation produced by humans?
- How can AI-generated disinformation be identified?
- How can large language models (LLMs) influence the perceived persuasiveness of disinformation?
The researchers will take as their point of departure datasets developed in the project “Fakespeak: The language of fake news”, which explored how linguistic cues can detect fake news. These datasets will be compared with the output from various popular LLMs.
The texts will be analysed from a linguistic and media studies perspective. In the next step, computer scientists will investigate how the new knowledge can contribute to developing tools for identifying AI-generated disinformation.
