Consent Management Platform (CMP)

Scans third-party vendor lists in real-time to classify risk based on data destination and processing intent, specifically identifying vendors involved in Large Language Model Operations (LLMOps).

• AI Vendor Taxonomy: Automatically flags vendors known to use data for secondary purposes like synthetic data generation or model training.
• Data Flow Mapping: Visualizes the chain of custody from the browser to the vendor's inference endpoint.
• Automated Blocking: Prevents firing tags from vendors classified as high-risk for derivative work detection bypass.

Synchronizes user privacy choices across multiple sub-domains and first-party contexts without relying on third-party cookies, maintaining a unified privacy posture.

• First-Party Storage: Utilizes a secure, HTTP-only cookie on a shared infrastructure domain to propagate preferences.
• JavaScript Bridge: Employs a non-blocking script to read the central consent state on sibling properties.
• AI Opt-Out Propagation: Ensures that an AI training opt-out on the main domain is instantly respected on the blog, docs, and support sub-domains.

Exposes consent decisions via a high-performance API, allowing server-side applications and Retrieval-Augmented Generation (RAG) pipelines to check permissions before data retrieval.

• Zero-Trust Integration: Before a vector database returns a chunk of text to an LLM, the system pings the CMP API to verify the user's RAG permissioning status.
• Microsecond Latency: Delivers boolean consent checks with sub-millisecond latency to avoid slowing down generation.
• Session Tokens: Issues short-lived tokens that carry the scope of consent for stateless backend services.

A browser-level signal that communicates a user's universal opt-out preference to every website they visit. When integrated with a CMP, the GPC signal is honored as a legally binding objection to data sale and sharing, including AI training ingestion.

• Transmitted via the Sec-GPC HTTP header
• Recognized under CCPA and Colorado Privacy Act as a valid opt-out mechanism
• Eliminates the need for per-site consent management
• A CMP must detect, log, and enforce GPC signals across all downstream data processors

A computational representation of the complete lifecycle of data, tracking its origin, transformations, and movement through AI pipelines. A CMP feeds consent metadata into the lineage graph to ensure continuous compliance.

• Tracks the moment consent was granted or revoked
• Propagates consent state changes to all downstream data stores
• Enables algorithmic disgorgement by identifying all models trained on non-consenting data
• Provides an immutable record for regulatory audits under the EU AI Act

A tamper-proof, chronological record of all consent collection, modification, and transmission events. This log serves as the legal proof of compliance required by data protection authorities.

• Records every user interaction with the CMP interface
• Logs the exact timestamp and version of the consent string transmitted
• Stored on append-only storage to prevent retroactive modification
• Essential for demonstrating compliance with GDPR Article 7 (conditions for consent)
• Supports defense against regulatory fines by proving the state of consent at any point in time