Asymmetric Reversibility After Reliance: Why Taking Back a Statement Does Not Undo What Others Did on the Strength of It

This conceptual paper introduces Asymmetric Reversibility After Reliance, a mechanism in which one party can retract a statement, promise, assurance, or plan after another party has already spent resources, made commitments, or closed alternatives in reliance on it. The account assesses withdrawal and restoration separately for each party. Retraction establishes that the speaker no longer stands by the initiating statement. It does not establish that the situation has returned to where it was before the statement was made. Some consequences of reliance may be reversible, while others may be costly or impossible to undo, and some reliance may have been premature or unreasonable. The discussion connects the mechanism to Fuller and Perdue’s account of the reliance interest and examines Hoffman v. Red Owl Stores (1965). The Hoffmans sold property, paid money, and changed their position in response to assurances about a proposed grocery franchise. Later revisions of the required investment did not reverse those earlier acts, and the Wisconsin Supreme Court treated the reliance as a ground for relief. Part of the Coordination and Dependence series within On Human Modeling, the paper examines how reversibility can differ across participants once one has acted on another’s statement. Its framework can inform human–AI research on revisable outputs, changing recommendations, agent commitments, and systems whose technically editable statements may nevertheless produce downstream human commitments that cannot be restored by simply changing the original output.