Victoria Vziatysheva

Mykola Makhortykh, Victoria Vziatysheva, Maryna Sydorova

The rise of generative artificial intelligence (genAI) models poses new possibilities and risks for how the past is remembered by accelerating content production and altering the process of information discovery. The most critical risk is historical misrepresentation, which ranges from the distortion of facts and inaccurate depiction of specific groups to more subtle forms, such as the selective moralization of history. The dangers of misrepresentation of the past are particularly pronounced for high-risk memories, such as memories of past atrocities, which have a strong emotional load and are often instrumentalised by political actors. To understand how substantive this risk is, we empirically investigate how genAI applications deal with high-risk memories of the Second World War atrocities in Ukraine. This case is crucial due to the scope of the atrocities and the intense, often instrumentalised, contestation surrounding their memory. We audit the performance of three common genAI applications for different types of misrepresentation, including hallucinations and inconsistent moralization, and discuss the implications for future memory practices.

Vihang Jumle, Mykola Makhortykh, Maryna Sydorova et al.

Framing is among the most extensively used concepts in the field of communication science. The availability of digital data offers new possibilities for studying how specific aspects of social reality are made more salient in online communication but also raises challenges related to the scaling of framing analysis and its adoption to new research areas (e.g. studying the impact of artificial intelligence-powered systems on representation of societally relevant issues). To address these challenges, we introduce a transformer-based approach for generic news frame detection in Anglophone online content. While doing so, we discuss the composition of the training and test datasets, the model architecture, and the validation of the approach and reflect on the possibilities and limitations of the automated detection of generic news frames.

Elizaveta Kuznetsova, Mykola Makhortykh, Victoria Vziatysheva et al.

This article presents a comparative analysis of the ability of two large language model (LLM)-based chatbots, ChatGPT and Bing Chat, recently rebranded to Microsoft Copilot, to detect veracity of political information. We use AI auditing methodology to investigate how chatbots evaluate true, false, and borderline statements on five topics: COVID-19, Russian aggression against Ukraine, the Holocaust, climate change, and LGBTQ+ related debates. We compare how the chatbots perform in high- and low-resource languages by using prompts in English, Russian, and Ukrainian. Furthermore, we explore the ability of chatbots to evaluate statements according to political communication concepts of disinformation, misinformation, and conspiracy theory, using definition-oriented prompts. We also systematically test how such evaluations are influenced by source bias which we model by attributing specific claims to various political and social actors. The results show high performance of ChatGPT for the baseline veracity evaluation task, with 72 percent of the cases evaluated correctly on average across languages without pre-training. Bing Chat performed worse with a 67 percent accuracy. We observe significant disparities in how chatbots evaluate prompts in high- and low-resource languages and how they adapt their evaluations to political communication concepts with ChatGPT providing more nuanced outputs than Bing Chat. Finally, we find that for some veracity detection-related tasks, the performance of chatbots varied depending on the topic of the statement or the source to which it is attributed. These findings highlight the potential of LLM-based chatbots in tackling different forms of false information in online environments, but also points to the substantial variation in terms of how such potential is realized due to specific factors, such as language of the prompt or the topic.

Elizaveta Kuznetsova, Ilaria Vitulano, Mykola Makhortykh et al.

The purpose of this study is to assess how large language models (LLMs) can be used for fact-checking and contribute to the broader debate on the use of automated means for veracity identification. To achieve this purpose, we use AI auditing methodology that systematically evaluates performance of five LLMs (ChatGPT 4, Llama 3 (70B), Llama 3.1 (405B), Claude 3.5 Sonnet, and Google Gemini) using prompts regarding a large set of statements fact-checked by professional journalists (16,513). Specifically, we use topic modeling and regression analysis to investigate which factors (e.g. topic of the prompt or the LLM type) affect evaluations of true, false, and mixed statements. Our findings reveal that while ChatGPT 4 and Google Gemini achieved higher accuracy than other models, overall performance across models remains modest. Notably, the results indicate that models are better at identifying false statements, especially on sensitive topics such as COVID-19, American political controversies, and social issues, suggesting possible guardrails that may enhance accuracy on these topics. The major implication of our findings is that there are significant challenges for using LLMs for factchecking, including significant variation in performance across different LLMs and unequal quality of outputs for specific topics which can be attributed to deficits of training data. Our research highlights the potential and limitations of LLMs in political fact-checking, suggesting potential avenues for further improvements in guardrails as well as fine-tuning.