"There is Nothing We Can Do" for Some and "I Can Do Everything" for Others : AI Revolution and Perceived Behavioral Control Across Social Classes in the D.R.Congo

Anticipating disruptive technologies often reveals deep social divides in how individuals interpret change. This paper examines the psychological consequences of imagining an Artificial Intelligence (AI) revolution in the Democratic Republic of Congo (DRC), with particular attention to perceived behavioral control (PBC). Drawing on the Theory of Planned Behavior, we conceptualize PBC through two key dimensions: self-efficacy and locus of control. Using a randomized experimental design with treatment, placebo, and control conditions, participants were primed with narratives of AI disruption, after which shifts in their sense of agency were measured. Findings indicate that AI primes significantly reduced perceptions of controllability, especially among disadvantaged groups, while self-efficacy remained largely stable. Conversely, individuals with relative advantages, proxied by car ownership and male gender, demonstrated resilience, and in certain cases even a rebound effect, suggesting that access to material and symbolic resources protects against the disempowering effects of disruptive change. These results underscore a critical psychological cleavage: whereas advantaged participants are inclined to view AI as opportunity, marginalized participants experience it as an uncontrollable threat. The study contributes to debates on inequality, perceived behavioral control, and technology adoption by revealing how social class moderates psychological responses to anticipated technological transformations. Policy implications emphasize that reducing inequality requires not only digital infrastructure and skill-building, but also the cultivation of psychological resources such as Locus of control, inclusion, and self-efficacy to mitigate the risk of AI amplifying existing divides.