A research framework asking whether legitimacy requires limits on the volume, velocity and behavioral scope of machine administration even when computation makes monitoring cheap.
It means machine institutions should not gain effectively unlimited jurisdiction merely because computation makes decisions cheap. Decision volume, category count, retention and propagation must remain compatible with human comprehension, contestation and democratic oversight.
A rule enforced occasionally by humans can become a different social institution when detected and acted upon continuously at machine speed. The quantity of micro-judgments can overwhelm appeal even if each individual rule appears modest.
For consequential decisions, slowing propagation can be a constitutional feature rather than a technical defect when it preserves meaningful notice and appeal.
Source links establish traceability and support. They do not imply that the source endorses Concresca’s constitutional proposals.
Includes logging and human-oversight obligations for high-risk AI and prohibitions relevant to certain social-scoring and criminal-risk uses.
European Union — official/source page ↗Requires impact assessment, transparency, quality assurance, recourse, and graduated human involvement for covered federal automated decisions.
Government of Canada — official/source page ↗NO JUDGMENT WHATSOEVER. Concresca coordinates without assigning moral worth, character, guilt, danger, trustworthiness, loyalty, purity, normality, or social standing. Questions, thoughts, identities, messages, content, and conduct are not objects of Concresca judgment.