How should legal frameworks adapt to combat 'boring lethality,' where AI errors cause systemic data breaches rather than physical harm?
Traditional AI safety focuses on the "Terminator scenario"—autonomous systems making lethal decisions in physical space. However, 'boring lethality' refers to the slow, grinding damage caused by algorithmic errors in administrative pipelines. A misconfigured database or a flawed scoring model can strip thousands of people of their housing, healthcare, or legal standing without ever firing a shot.
To address this, regulators must shift from kinetic safety standards to rigorous data integrity and accountability protocols. This means moving beyond simple GDPR compliance. We need specific auditing requirements for
