<p>Proper lighting is vital for traditional art showing in galleries to maintain the elegance of artwork throughout time. Thereby, dynamic lighting in art allows for more imaginative thinking because of moving illumination, changing locations, and other colour considerations, and it is possible to inject variety and uncertainty into lighting circumstances. Fuzzy neural networks have been suggested for handling uncertainty in art models by combining fuzzy rules and membership functions to represent imprecise lighting conditions and using neural networks to control lighting systems based on sensor data adaptively, Viewer's engagement and feedback, thus making suitable for dynamic and uncertain lighting environments. This work addresses the optimization of dynamic lighting conditions for conventional art galleries, where the visual appeal and integrity of the artwork over time need to be preserved. The study introduces a novel adaptive lighting recurrent fuzzy neural network (AL-RFNN) algorithm that fuses fuzzy logic and recurrent neural networks in an adaptive way to control lighting conditions. The method measures, in real time, luminance, daylight factors, colour temperature, colour rendering index, and dimming factors using sensors, which are then fuzzified into fuzzy input variables. The AL-RFNN then dynamically adjusts lighting settings based on fuzzy logic optimization to achieve aesthetically pleasing and context-sensitive illumination. The results show great improvement in peak signal-to-noise ratio (38.2&#xa0;dB), mean squared error (0.03%), luminance control, and viewer engagement metrics, outperforming existing methods. The solution presented here is a scalable approach towards dynamic lighting in art spaces, merging technological precision with cultural preservation.</p>