On a dark morning in March 2018, a self-driving Uber vehicle struck and killed Elaine Herzberg in Tempe, Arizona—the first fatal collision involving an autonomous vehicle. The incident exposed a chasm in AI accountability that remains largely unresolved today. When the vehicle's safety systems failed and the human backup driver did not intervene, the question of responsibility fractured across multiple actors: Was it the engineer who wrote the perception algorithm? The safety driver who wasn't paying attention? Uber's leadership who greenlit public testing? Or regulators who hadn't established clear oversight? The lack of a clear answer revealed that AI governance had outpaced legal and organizational frameworks designed to assign liability. Uber eventually settled with Herzberg's family, but the case crystallized a broader truth: companies deploying AI systems must urgently define who owns outcomes when algorithms fail, or face cascading regulatory and reputational consequences.