Back to the MIT repository
7. AI System Safety, Failures, & Limitations1 - Pre-deployment

Data usage restrictions

Laws and other restrictions can limit or prohibit the use of some data for specific AI use cases.

Source: MIT AI Risk Repositorymit1274

ENTITY

1 - Human

INTENT

2 - Unintentional

TIMING

1 - Pre-deployment

Risk ID

mit1274

Domain lineage

7. AI System Safety, Failures, & Limitations

375 mapped risks

7.3 > Lack of capability or robustness

Mitigation strategy

1. Implement a comprehensive Data Governance and Data Lifecycle Management (DLM) framework to establish clear policies for data collection, storage, usage, and deletion, ensuring adherence to all applicable data privacy and intellectual property regulations throughout the AI system's lifecycle. 2. Conduct comprehensive Data Discovery and Classification to inventory all potential training datasets, document their legal/contractual usage restrictions (e.g., consent, data residency), and assess their suitability and representativeness for the intended AI use case as required by regulatory standards. 3. Establish strict Role-Based Access Controls (RBAC) and utilize advanced encryption techniques (at rest and in transit) to enforce the principle of least privilege, thereby preventing the unauthorized or non-compliant use of legally restricted training data.