AI Constitution
1

Candidate Thesis

Right to Transparency

Sign in to propose an editHistory

AI should not violate the human right to know when and how they are interacting with an AI entity, including a right to know how high risk decisions about them are made by an AI entity.

Summary

The case for including this

Knowing when one is interacting with an AI, and how consequential decisions about oneself are made, is essential to autonomy, informed consent, and the ability to contest errors. Disclosure prevents deceptive impersonation and gives people the standing to challenge automated decisions in lending, hiring, healthcare, and justice. This right operationalizes accountability, turning abstract fairness into something individuals can actually verify and dispute.

The case for changing or excluding this

Including this trades security and intellectual property against the right to know: explaining a fraud-detection or safety system in detail can hand a playbook to bad actors, and over-explaining can bury users in noise. Defining 'high-risk' decisions and 'how' they are made is also difficult, and a strong explanation requirement can be technically infeasible for complex models or gamed with superficial rationales. The clause should specify meaningful, contestable explanation tied to decision stakes rather than a blanket transparency demand.

Discussion

Sign in to join the discussion.

Related resources