Disinformation Capabilities of Large Language Models

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F24%3APU154648" target="_blank" >RIV/00216305:26230/24:PU154648 - isvavai.cz</a>

Result on the web

<a href="https://aclanthology.org/2024.acl-long.793" target="_blank" >https://aclanthology.org/2024.acl-long.793</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Disinformation Capabilities of Large Language Models

Original language description

Automated disinformation generation is often listed as one of the risks of large language models (LLMs). The theoretical ability to flood the information space with disinformation content might have dramatic consequences for democratic societies around the world. This paper presents a comprehensive study of the disinformation capabilities of the current generation of LLMs to generate false news articles in English language. In our study, we evaluated the capabilities of 10 LLMs using 20 disinformation narratives. We evaluated several aspects of the LLMs: how well they are at generating news articles, how strongly they tend to agree or disagree with the disinformation narratives, how often they generate safety warnings, etc. We also evaluated the abilities of detection models to detect these articles as LLM-generated. We conclude that LLMs are able to generate convincing news articles that agree with dangerous disinformation narratives.

Czech name

—

Czech description

—

Type

O - Miscellaneous

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year

2024

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů