The article presents a linguistic analysis of the mechanisms underlying viral content dissemination in an algorithmically governed digital media environment. The relevance of the study is determined by the ongoing transformation of public communication on digital platforms, where the selection and amplification of messages are increasingly driven by formalized engagement metrics rather than interpersonal social ties. This shift necessitates a reconsideration of traditional approaches to the analysis of public discourse and the development of conceptual tools that account for algorithmic mediation. The research problem lies in identifying stable linguistic characteristics of algorithmically amplified content and in describing the structure of the algorithmic idiolect – a specific speech register adapted to the criteria of platform-based selection and ranking. Particular attention is paid to the relationship between formal linguistic strategies, discursive organization of messages, and the metrics monitored by platform algorithms. The study employs an interdisciplinary approach combining methods from media linguistics, communication theory, and applied data analysis. The empirical basis consists of a corpus of 2,000 social media posts published between January and June 2025, divided into a viral subset (1,000 posts with the highest engagement indicators) and a control subset (1,000 posts with average engagement). Quantitative methods were applied, including calculation of average sentence length, frequency of expressive constructions, automated sentiment analysis using a neural network model, and content analysis of pragmatic and multimodal elements. The statistical significance of the observed differences was tested using a Student's t-test and a chi-squared test. The results demonstrate that viral content is characterized by syntactic reduction, increased lexical expressivity, polarity of sentiment, reliance on binary oppositions, and a high degree of multimodality, which significantly enhances user engagement. Based on the identified features, the study substantiates the notion of an algorithmic idiolect of virality as a standardized linguistic register of digital communication optimized for platform algorithms.