This study examines how algorithmic bias (AB), deepfake technology (DT), and AI-generated advertising bias (AIAB) influence consumer trust in digital messages, with a focus on the mediating roles of perception of authenticity (POA) and cognitive response (CR), and the moderating role of technological literacy (TL) in Ghana’s digital advertising context. Using a hybrid PLS-SEM and artificial neural network (ANN) approach, data from 510 e-commerce users revealed that AB and DT reduce POA, while AIAB negatively affects CR. Both POA and CR positively influence trust in messages, and TL weakens the negative effects of DT and AIAB, indicating that digitally literate consumers are more discerning. The study contributes to theory by contextualizing algorithmic influence in Sub-Saharan Africa and highlights policy implications for promoting digital literacy and algorithmic transparency