Clinical AI Applications in the Context of Cancer: Explaining Openness to Information and Consent to Use Through Risk Perceptions and Efficacy Beliefs

The use of artificial intelligence (AI) in oncology holds immense potential for improving diagnostic accuracy and treatment outcomes. Yet, concerns about potential risks need to be addressed. To improve acceptance, understanding, and trust, it is crucial to provide individuals with accurate information about AI in medicine. Building on the Risk Perception Attitude Framework (RPA), this study examines how combinations of AI-related risk perceptions and AI-information efficacy beliefs shape individuals' openness to information and consent to the use of AI-assisted cancer care. Data from an online survey of the German population stratified by age, gender, and education (<i>n</i> = 3,274) were segmented into four groups based on different patterns of AI-related risk perceptions and AI-information efficacy beliefs using the RPA framework. Results show significant differences in information-seeking intentions and willingness to consent to the use of AI in oncology care across groups. Our study extends the RPA framework to technological innovations in medicine and offers implications for targeted strategies for health communication campaigns aimed at improving acceptance of AI use in oncology.