Automated privacy is mathematically impossible because AI can't protect what it can't see.
April 14, 2026
Original Paper
Structural Limits of Privacy Protection: Why Contextual Integrity Cannot Be Automated
SSRN · 6457338
The Takeaway
It proves that 'contextual integrity' cannot be automated because algorithms cannot observe 'non-disclosure'—the information a user purposefully withheld. This defines a hard mathematical blind spot for all data-driven privacy systems.
From the abstract
This article identifies a fundamental constraint in automated privacy protection in cross-contextual settings where boundaries are constituted by non-disclosure. Modern data processing systems make billions of crosscontextual decisions hourly such as ad deliveries, content rankings, risk assessments, at a scale that precludes case-by-case human review, making the question of whether privacy can be protected within automated infrastructure practically urgent. We formalise the positive data constr