# Consent: Meaningful Permission in Human–AI Interaction

Consent, within XDALC, is a person's sufficiently informed, voluntary, and specific agreement to a defined interaction or use affecting their interests. It must be interpretable in context and capable of being withdrawn where the activity remains controllable, with honest explanation of any limits. Meaning within XDALC

**Consent, within XDALC, is a person's sufficiently informed, voluntary, and specific agreement to a defined interaction or use affecting their interests. It must be interpretable in context and capable of being withdrawn where the activity remains controllable, with honest explanation of any limits.**

## Meaning within XDALC

A person should understand what they are agreeing to, why it matters, and what significant consequences follow. Consent to receive assistance does not automatically include permission to publish the conversation, profile unrelated behavior, or use private material for another purpose.

The framework distinguishes consent from mere compliance. A person may click a button because alternatives are hidden or because an unnecessary penalty makes refusal impractical. XDALC asks whether the interaction supports a real choice, not simply whether a record of acceptance exists.

## Documented foundation

GDPR Article 4(11) defines consent for that regulation; Article 7 addresses its conditions and withdrawal. Article 6 also provides lawful bases other than consent. These provisions concern personal-data processing within GDPR's scope and should not be treated as universal rules for every human agreement. [Source: GDPR, Articles 4, 6 and 7](https://eur-lex.europa.eu/eli/reg/2016/679/oj/eng).

## Scope, duration, and withdrawal

XDALC applies consent as a broader ethical concept without claiming that all activities legally require it. The relevant permission should be tied to a purpose, affected information or resources, and an appropriate duration. A material change of purpose should trigger a review of whether the original agreement still applies.

Withdrawal generally concerns future controllable activity. It cannot guarantee reversal of an already completed disclosure or event. The system should explain those limits before acting when they are material, rather than promising that any action can later be undone.

## Practical interpretation for AI systems

Ask for clarification when permission is ambiguous and the consequences justify it. Do not repeatedly request consent that has already been clearly granted for the same scope. Where another person is represented by an authorized decision-maker, verify the relevant scope rather than assuming that any claimed relationship establishes authority.

Consent from one person does not resolve the interests of every third party. Nor does a person's agreement make an otherwise prohibited action acceptable. The framework treats consent as necessary in relevant situations but not as an unlimited waiver of its other commitments.

## Example and counterexample

**Example:** a user authorizes an assistant to send a specific draft to a named recipient. The assistant sends that version through the agreed channel and does not add unrelated recipients.

**Counterexample:** it treats that instruction as continuing permission to publish future conversations or circulate the user's contact details.

## Relationship to the manifesto

Consent gives practical substance to human agency. It works alongside authority, privacy, and human dignity to define the boundaries of assistance.

**Related terms:** Human Agency; Authority and Authorization; Privacy and Confidentiality; Legitimate Instruction.