In the digital age, traditional moral categories undergo significant transformation, both in meaning and in the ways they are expressed. Mercy, deeply rooted in philosophical and religious traditions, has become a subject of mass discussion in social media, where it acquires new forms of representation. The relevance of this study stems from the need to empirically explore how values traditionally associated with ethics and axiology are interpreted and reproduced in user practices of online interaction. The novelty of the research consists in analyzing mercy not in an abstract philosophical framework but through a large corpus of user-generated content from the Russian segment of social media collected between September 30, 2024 and April 1, 2025. Using the Brand Analytics system, over one million public messages were retrieved; topic modeling (LDA) revealed 25 clusters of values, among which the “Kindness and Humanity” cluster (approximately 70,000 messages) was of primary relevance. The focused analysis was conducted on a sub-sample of about 19,000 texts explicitly containing mercy-related lexicon. The results demonstrate that mercy most frequently appears in contexts of requests for help and charitable practices – both toward people and animals – along with congratulatory and gratitude formulas, personal narratives and reflections, and its linkage with demands for justice. The distribution across contexts is analytical rather than rigid, aiming to show the relative volume of messages while acknowledging possible overlaps and interpretative variations. The findings confirm the ambivalent nature of mercy: on the one hand, it retains the status of a sacralized moral ideal; on the other, it functions as an instrumental frame for mobilizing resources and as an element of digital self-presentation. Overall, mercy in social media emerges as a hybrid value, combining spiritual and ethical orientations with utilitarian practices of collective action and social bonding.