Abstract
The modern era of rapid artificial intelligence development witnessed a fundamental paradigmatic shift in human-computer interaction – from technocratic manipulation to conscious expression through natural language. The study aimed to philosophically examine the transition from manipulative to expressive interaction with natural language interfaces. An interdisciplinary approach was employed, combining methods of philosophical analysis, cognitive phenomenology, and digital culture critique. The findings revealed that modern large language model-based interfaces not only simplified communication but also reconfigured the ontological foundations of technological experience. The growing “semantic opacity” of large language systems challenged traditional notions of understanding, accountability, and explainability in technical interaction. A phenomenological shift toward “invisible” technology integrated into the subject’s cognitive architecture was identified. The phenomenon of technological intersubjectivity was analysed, blurring the boundary between human and machine agency. It was demonstrated that language as an interface not only transmitted information but also functioned as a medium for shaping self-identity, autonomy, and cognitive responsibility. The conclusions proved that this transformation carried profound philosophical implications and required systematic reflection to prevent the erosion of human autonomy in the context of cognitive fusion with artificial intelligence. The necessity of an ethical framework for new interfaces was emphasised – one that ensured not only functionality but also the preservation of human agency. The practical significance lay in establishing philosophical foundations for designing natural language interfaces that supported human autonomy, responsibility, and epistemic independence