Upholding Epistemic Agency: A Brouwerian Assertibility Constraint for Responsible AI

Researchers propose a Brouwerian assertibility constraint for AI systems that requires them to provide publicly inspectable certificates of entitlement before making claims in high-stakes domains. The framework introduces a three-status interface (Asserted, Denied, Undetermined) to preserve human epistemic agency when AI systems participate in public justification processes.

• →Proposes requiring AI systems to provide verifiable certificates before making authoritative claims in critical domains
• →Introduces three-status output system (Asserted, Denied, Undetermined) instead of binary responses
• →Aims to preserve human decision-making agency by making AI outputs contestable and transparent
• →Uses mathematical framework inspired by Brouwerian logic to ensure AI accountability
• →Addresses concern that generative AI can convert uncertainty into seemingly authoritative verdicts