AI literacy and school tier moderate educators' ethical coping mechanisms through sedative and reverse Matthew effects.

The rapid penetration of artificial intelligence in education has triggered unprecedented ethical risk perception and response challenges for educators, but existing research lacks systematic exploration of deep ethical cognitive mechanisms. To address this, this study operationalizes macro level sociotechnical imaginaries into micro level psychological perceptions to construct and test a theoretical model that integrates dual-dimensional risk perception, AI Literacy, and situational factors. PLS-SEM analysis of 502 Chinese educators reveals that: both Inhibitory Ethical Risk and Promotional Professional Risk are significantly and positively associated with Defensive and Constructive Coping strategies simultaneously, challenging the traditional one dimensional risk hypothesis; AI Literacy significantly attenuated the positive association between Inhibitory Ethical Risk and Constructive Coping, supporting the hypothesis of the Literacy Sedative Effect; The strength of the positive association between Promotional Professional Risk and Constructive Coping in lower tier schools is 2.78 times stronger than that in first tier schools, providing empirical evidence for the Reverse Matthew Effect. This study proposes a dual-dimensional risk-coping framework, revealing the cognitive decoupling mechanism of AI Literacy and challenging the classical Matthew Effect, providing a new perspective for understanding individual behavior in disruptive technological change; at the same time, it also provides practical guidance for cultivating cautious and proactive adopters and developing differentiated AI Literacy training strategies.