Artificial intelligence is currently one of the most controversial discussed technologies across various work domains. In healthcare, AI fosters widespread positive beliefs of substantially increasing the quality of care, yet evoking physicians’ fears of being marginalized or replaced. The described controversy leads to ambivalent attitudes, as physicians hold both strong positive and negative evaluations at the same time. However, current research in information systems has not been able to measure ambivalence because uni-polar attitude scales cannot assess this construct. Additionally, it is unclear whether ambivalence has positive or negative consequences and how it is related to resistance to change. In the context of AI in healthcare, we conducted a survey study (n=74) to measure context-specific attitudes and attitude ambivalence of physicians. We distinguish between two states of ambivalence and show that physicians who experience an inner conflict (Felt Ambivalence) from conflicting attitudes (Potential Ambivalence) develop resistance to change. Moreover, including ambivalence into a regression model explains more variance than uni-polar attitudes alone. With this research, we show that ambivalent attitudes can be measured in the context of technological change. Additionally, we explore how context-specific attitudes towards AI in healthcare drive physicians’ ambivalence towards it.
Maier, Sophia Bettina; Jussupow, Ekaterina; and Heinzl, Armin, (2019). "GOOD, BAD, OR BOTH? MEASUREMENT OF PHYSICIAN’S AMBIVALENT ATTITUDES TOWARDS AI". In Proceedings of the 27th European Conference on Information Systems (ECIS), Stockholm & Uppsala, Sweden, June 8-14, 2019. ISBN 978-1-7336325-0-8 Research Papers.