This preprint introduces Crisis Amplification, a platform-optimized behavioral pattern in contemporary social-media political discourse. Through qualitative digital ethnography, the study demonstrates how heterogeneous events are systematically framed as urgent moral emergencies to maximize engagement under attention-driven incentive structures. Unlike misinformation models, Crisis Amplification does not rely on falsity or deceptive intent; instead, it exploits temporal asymmetries between engagement and correction. Corrections occur post-amplification and lack equivalent algorithmic velocity, rendering them structurally ineffective. The analysis reframes outrage-driven political communication as a predictable outcome of platform design rather than individual pathology, with implications for digital governance, moderation strategies, and epistemic resilience.

This work is a qualitative, theory-oriented preprint intended to formalize a recurrent behavioral pattern observed across large social platforms. It avoids intent attribution and partisan positioning, focusing instead on incentive-driven dynamics that generalize across ideologies.