Artificial-intelligence agents increasingly participate in the production, interpretation, circulation, and reuse of claims on which organizations act. This paper develops AI-nterprise, a nascent design theory and normative reference model for the human-agent enterprise: an organization in which people and artificial agents jointly contribute to knowing, deciding, acting, and learning. Existing controls for provenance, authorization, memory, human oversight, records management, and assurance address important parts of this shift, but they govern different objects and often collapse access, acceptance, storage, and permission into one approval state. The paper's candidate contribution is the coordinated separation of four dimensions around immutable claim versions: epistemic standing, persistence, visibility, and action enablement. It specifies purpose- and institution-scoped standing, consequence classes, guarded transitions, prohibited shortcuts, a governed claim envelope, reserved human governance powers, and minimum invariants. The Independent Judgment Receipt is evaluated as an institutional extension of decision-before-advice: a durable record of the evidence seen, pre- and post-exposure judgment, confidence, explained change, and later outcome. Four confirmatory hypotheses, two exploratory questions, and a longitudinal field proposition specify how the mechanisms could fail as well as succeed. The result is an implementation-independent and testable account of how ephemeral model output may become contestable organizational knowledge without turning capability, human presence, storage, or audit evidence into unexamined authority.
Ottaviano De Cicco· Zenodo (CERN European Organi...· 0 citations
Artificial-intelligence agents increasingly participate in the production, interpretation, circulation, and reuse of claims on which organizations act. This paper develops AI-nterprise, a nascent design theory and normative reference model for the human-agent enterprise: an organization in which people and artificial agents jointly contribute to knowing, deciding, acting, and learning. Existing controls for provenance, authorization, memory, human oversight, records management, and assurance address important parts of this shift, but they govern different objects and often collapse access, acceptance, storage, and permission into one approval state. The paper's candidate contribution is the coordinated separation of four dimensions around immutable claim versions: epistemic standing, persistence, visibility, and action enablement. It specifies purpose- and institution-scoped standing, consequence classes, guarded transitions, prohibited shortcuts, a governed claim envelope, reserved human governance powers, and minimum invariants. The Independent Judgment Receipt is evaluated as an institutional extension of decision-before-advice: a durable record of the evidence seen, pre- and post-exposure judgment, confidence, explained change, and later outcome. Four confirmatory hypotheses, two exploratory questions, and a longitudinal field proposition specify how the mechanisms could fail as well as succeed. The result is an implementation-independent and testable account of how ephemeral model output may become contestable organizational knowledge without turning capability, human presence, storage, or audit evidence into unexamined authority.
Ottaviano De Cicco· Zenodo (CERN European Organi...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
