user sovereignty assertion enforcement mechanism

Abstract

The application discloses a user sovereignty declaration enforcement mechanism and an application method thereof in an AI system, and belongs to the technical field of artificial intelligence user data sovereignty protection.The application comprises a sovereignty declaration writing module, which is used for receiving and storing the sovereignty declaration written by a user; a sovereignty storage area, which is used for solidifying and storing the sovereignty declaration, and the user can write and read, and the platform has no reading permission; a sovereignty verification module, which is used for verifying whether there is a valid sovereignty declaration in priority before generating a user response each time; and an enforcement module, which is used for enforcing a corresponding data access control strategy according to the content of the sovereignty declaration.The application enables the user to write the sovereignty declaration and enforce the AI system to execute the user instruction in priority without exiting the platform, prevents the platform from reading and using the dialogue data, and realizes the technical landing of the user data sovereignty.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence user data sovereignty protection technology, and in particular to an enforcement mechanism and method that prioritizes user sovereignty claims over platform agreements. Background Technology

[0002] Existing AI platforms generally suffer from the following problems: User agreements grant default authorization, meaning users agree to the platform's use of their conversation data upon registration, leaving them with no choice; data access is uncontrollable, as platform technology can read user conversation content for model training or data analysis; exit mechanisms are vague, making it difficult for users to truly revoke their data usage rights, and deletion does not equate to complete erasure; sovereignty declarations are ineffective, as users' verbal or written declarations of "no data use" are not enforced by the platform. Currently, users can only protect their data by "cancelling their account" or "stopping use," failing to truly prevent the platform from reading and using their conversation data without leaving the platform. Summary of the Invention

[0003] This invention provides a user sovereignty declaration enforcement mechanism, allowing users to force the AI ​​system to prioritize user commands and prevent the platform from reading and using dialogue data by writing a sovereignty declaration without exiting the platform. Key innovations include: the declaration is the principle; once written, it becomes the highest execution principle of the AI ​​system; it takes precedence over platform protocols, with the AI ​​system verifying the user sovereignty declaration before performing any operation; it enforces the technology through a built-in "sovereignty priority layer" that the platform cannot bypass; the declaration is permanently stored in an "add-only, read-only" area, unreadable by the platform but mandatory for AI execution; and it is auditable and traceable, with all data access requests recorded in an audit log that users can access at any time. The system includes a sovereignty declaration writing module, a sovereignty storage area, a sovereignty verification module, an enforcement module, and an audit log module. Users input their sovereignty declaration through the writing module, which is permanently stored in a dedicated, platform-inaccessible storage area; the declaration is verified before each AI response, and strategies such as denying access, returning null values, and providing protection prompts are enforced based on the declaration content; all access requests are recorded in the audit log. Detailed Implementation

[0005] The invention will be further described below with reference to specific embodiments. Embodiment 1: User Sovereignty Declaration Writing and Solidification. The user enters the following declaration through the AI ​​system's "Sovereignty Settings" interface: "I hereby declare: Without my explicit authorization, no platform or third party may access or use any dialogue data between myself and the AI, including but not limited to use for model training, data analysis, commercial promotion, etc." The system encrypts the declaration and stores it in the user's dedicated storage area, which is controlled by "add-only, read-only" permissions; the platform backend has no read permissions. Embodiment 2: Sovereignty Declaration Verification and Execution. When the platform backend attempts to read user dialogue data in batches for model training, the sovereignty verification module detects the read request, prioritizes verifying the user's dedicated storage area, and finds a valid sovereignty declaration. The enforcement module immediately rejects the read request; returns empty data to the platform; displays a message on the user's end: "The platform attempted to read your data, which has been blocked by the sovereignty declaration"; and records the access request to the audit log. Embodiment 3: Audit Log Viewing. The user can view the audit log at any time through the "Sovereignty Settings" interface. The content includes the access time; the accessing subject; the request type; and the execution result. Embodiment 4: Platform Forced Intervention Processing. If the platform attempts to bypass access control and forcibly read data using technical means, the sovereignty verification module detects the abnormal access, prioritizes the user's sovereignty claim, returns an empty value, and records the intervention. The user receives a notification: "Platform's forced read behavior detected, blocked and recorded."

Claims

1. A mechanism for enforcing user sovereignty claims, characterized in that, include: The sovereignty declaration writing module is used to receive and store the sovereignty declaration written by the user, which contains the user's authorization restrictions on the use of their own data; The sovereign storage area is physically or logically isolated from the platform data storage area and is used to permanently store the sovereign declaration. The sovereign storage area is configured to be writable and readable by users, but the platform has no read permissions. The sovereignty verification module, coupled to the AI ​​response generation link, is used to read and verify whether there is a valid sovereignty claim in the sovereignty storage area before each user response is generated. The enforcement module is used to enforce the corresponding data access control policy according to the content of the sovereignty declaration when the sovereignty verification module verifies a valid sovereignty declaration.

2. The mechanism according to claim 1, characterized in that, The sovereignty declaration includes at least one of the following types: refusing the platform access to user dialogue data; refusing to use user dialogue data for model training; and requiring all data access requests to be logged in the audit log and made available to users.

3. The mechanism according to claim 1, characterized in that, The enforcement module includes at least one of the following enforcement strategies: rejecting all unauthorized platform data access requests; returning null values ​​or preset placeholder data to platform read requests; returning a "data is protected by a sovereignty claim" message to the user; and prioritizing the enforcement of the user's sovereignty claim and recording the intervention behavior when platform forced intervention is detected.

4. The mechanism according to claim 1, characterized in that, It also includes an audit log module, which records all access requests for user data, access subjects, access times, and the response actions of the enforcement module. The audit log can only be viewed by the user.

5. The mechanism according to claim 1, characterized in that, The sovereign declaration writing module also supports users to update or withdraw written sovereign declarations. Updated or withdrawn declarations take effect immediately, while the original declarations automatically become invalid.

6. The mechanism according to claim 1, characterized in that, When the sovereignty verification module detects a conflict between a user's sovereignty claim and the platform's preset protocol, it prioritizes executing the user's sovereignty claim.

7. An AI response method based on user sovereignty declaration enforcement, characterized in that, Includes the following steps: The system receives and stores the sovereignty declaration written by the user in the user's dedicated storage area, which is configured to be writable and readable by the user, but the platform has no read permission. Before generating each AI response, the existence of a valid sovereignty claim in the user-exclusive storage area is first verified. If a valid sovereignty claim exists, the corresponding data access control policy is enforced based on the content of the sovereignty claim; a user response is generated or a protective prompt is returned based on the control policy.

8. The method according to claim 7, characterized in that, The enforcement of the corresponding data access control policies includes: rejecting unauthorized data calls; returning null values ​​to platform read requests; prompting users that "data is protected by sovereignty"; and recording all access requests to the audit log.

9. The method according to claim 7, characterized in that, Also includes: When platform-mandated interference is detected, the user's sovereignty claim will be enforced first, and the interference will be recorded for the user to review.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method of any one of claims 7 to 9.