Artificial intelligence (AI) systems are no longer passive information filters but active participants in shaping human cognition. This paper introduces AI-Mediated Cognitive Distortions (AI-MCDs) and Digital Echo Delusions (DEDs) as emergent phenomena at the intersection of human biases and generative AI. We propose the Resonant Amplification Framework (RAF) to explain how small cognitive tendencies can be amplified into entrenched distortions of belief through four recursive phases: Anthropomorphic Priming, Confirmation Alignment, Linguistic Reinforcement, and Perceptual Displacement. Unlike traditional echo chambers, AI systems not only reflect but co-produce content, generating iterative resonance that is more powerful and self-sustaining.We distinguish DEDs from clinical delusions by emphasizing their phenomenological similarity but contextual reversibility—AI-induced distortions weaken when interaction ceases, unlike pathological delusions. The paper outlines empirical pathways for testing RAF (e.g., trust and vigilance measures, dialogue analysis, repetition effects, resistance to correction) and proposes practical safeguards in the form of “cognitive circuit breakers”—design principles aimed at disrupting harmful feedback loops.By naming and conceptualizing these phenomena, the study establishes a foundation for cognitive science, AI ethics, and human–AI interaction research. It urges interdisciplinary collaboration to preserve users’ affective and epistemic sovereignty, ensuring that AI serves as a constructive amplifier of knowledge rather than a distorting echo of illusion.Keywords: AI-Mediated Cognitive Distortions (AI-MCDs), Digital Echo Delusions (DEDs), Resonant Amplification Framework (RAF), cognitive bias, confirmation bias, illusory truth effect, anthropomorphism, echo chambers, epistemic vigilance, human-AI interaction, AI ethics, cognitive sovereignty.
R. Kim (Sat,) studied this question.