Algorithmic nudge to make better choices: Evaluating effectiveness of XAI frameworks to reveal biases in algorithmic decision making to users

Explainable AI (XAI) can be used as a practical tool to help users recognize and resist algorithmic bias, rather than just for technical transparency. The paper proposes using XAI frameworks to design interventions that alert users when algorithms are directing them toward problematic content like anti-vaccine videos.

Two types of explanations can support better decision-making: warnings about risks and counterfactual explanations showing alternative actions. The research distinguishes between "fore warnings" (which alert users to potential harms) and "counterfactual explanations" (which show users what they could do differently to change algorithmic outcomes), suggesting both approaches have value as decision aids.

Algorithmic explanations work best when paired with factual information about why content was recommended. Rather than explanations alone, the intervention design combines facts about algorithmic justification with either warnings or counterfactual guidance, suggesting that context matters for user understanding.

This approach treats users as active decision-makers rather than passive consumers. By revealing how algorithms work and offering concrete alternatives, the framework aims to empower users to make informed choices rather than simply accepting algorithmic recommendations at face value.