In light of global crises such as COVID-19, we argue that people’s trust in science drives their media choices in the digital sphere. The results from a German online survey show that individuals who perceive scientists as trustworthy in terms of ability, benevolence, and integrity, confidently expect science to provide accurate knowledge and guide reasonable decision-making in times of crisis. In turn, these positive trusting expectations towards science increase the use of established journalistic and scientific online sources for accessing scientific information. In contrast, people with low or no trust in science tend to resort to the messenger Telegram and ‘alternative’ online media outlets. Interestingly, the individual need for orientation does not amplify this association between trust in science and digital media use in a crisis situation.

Peter Weingart and Lars Guenther have written a short but nevertheless comprehensive stock-taking of science communication and the issue of trust. I fully agree with almost all of their theoretical and critical observations. My aim is to critically discuss the understanding of trust as expressed in the traditional discourse on science communication. From my point of view, this concept of trust in science reveals severe shortcomings. As a consequence, communication strategies following this concept could even jeopardize trust in science.

We argue that the institutionalized push communication of academic institutions has become the dominant form of public science communication and has tended to force other forms and functions of science communication into the background. Given the new schemes of public funding, public communication of science now primarily serves the purpose of enabling academic institutions to promote themselves in a competition that has been forced upon them by the political domain. What academics working under these conditions say about themselves and their work (and what they do not) will depend crucially on the strategic communication goals and concepts of the organizations to which they belong. We surmise that the inherent logic of this form of science communication represents a potential threat to the autonomy of scientific research.