Provision Registry

262 classified provisions across 299 platforms — browse, filter, and compare.

Every clause classified by type, severity, and platform. Updated as policies change.

Start Compliance free trial Track specific clauses across platforms with provision-level alerts.
Filtering: Ai automated × Clear all
medium Ai automated
Khan Academy · Khan Academy Terms of Service
The provision creates explicit boundaries on the service's intended use and establishes Khan Academy's position regarding responsibility for AI output quality. This framing allocates to users the responsibility to assess Khanmigo's suitability for their specific informational needs and to seek professional counsel where applicable.
CA-P-002870 First tracked Apr 18, 2026 Last seen Apr 18, 2026 Compare across platforms →
TikTok · TikTok Privacy Policy
This clause establishes the operational basis for TikTok's use of user-generated data and behavioral signals as inputs for artificial intelligence and machine learning development. The authorization covers both the collection mechanism (monitoring activity, scanning content) and the application of that data (model training, algorithm testing, technology improvement).
CA-P-008657 First tracked May 10, 2026 Last seen May 11, 2026 Compare across platforms →
OpenAI · GPT-4o System Card (PDF)
The provision establishes OpenAI's operational framework for ensuring predictable model behavior in production systems. Consistency mechanisms are significant for service reliability, user experience standardization, and the model's ability to perform its intended functions across varied use cases.
CA-P-000069 First tracked Apr 3, 2026 Last seen Apr 10, 2026 Compare across platforms →
Anthropic · Anthropic Privacy Policy
The provision establishes a default data use authorization for model training with conditional opt-out capability, while preserving Anthropic's ability to use certain categories of user content regardless of opt-out status. This structure means safety-flagged conversations and user-reported materials remain available for model improvement even when users exercise the opt-out.
CA-P-008335 First tracked May 10, 2026 Last seen May 11, 2026 Compare across platforms →
Cohere · Cohere Enterprise Data Commitments
This commitment has operational significance because it defines the scope of permitted data use for model development. The restriction creates a distinct data handling framework for enterprise customers compared to other service tiers.
CA-P-010605 First tracked May 11, 2026 Last seen May 11, 2026 Compare across platforms →
AWS Bedrock · AWS Service Terms
This clause defines a specific boundary on AWS's permitted uses of customer data within the Bedrock service. It establishes that model improvement through customer content training is excluded from AWS's operational scope for this service.
CA-P-008310 First tracked May 10, 2026 Last seen May 11, 2026 Compare across platforms →
Microsoft · Responsible AI
This provision establishes the organizational structure and assigns responsibility for developing and implementing responsible AI governance frameworks. The Office functions as the institutional mechanism for translating responsible AI commitments into operational practice across Microsoft's business units.
CA-P-007093 First tracked May 9, 2026 Last seen May 9, 2026 Compare across platforms →
OpenAI · GPT-4o System Card (PDF)
This architecture allows OpenAI to implement behavioral changes unilaterally and instantaneously across all users, creating a governance mechanism that operates outside traditional notice-and-consent or versioning frameworks. The operational significance is that system behavior can shift without triggering standard terms modification processes or user acknowledgment requirements.
CA-P-000068 First tracked Apr 3, 2026 Last seen Apr 10, 2026 Compare across platforms →
OpenAI · GPT-4o System Card (PDF)
The clause creates a gating mechanism that ties model advancement and release decisions to quantified risk assessment outcomes, establishing operational constraints on which model versions OpenAI may make available or continue iterating on.
CA-P-000066 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Microsoft · Responsible AI
This provision frames Microsoft's organizational commitment to AI governance across multiple dimensions. The inclusion of privacy and security as a stated principle establishes a documented standard against which the company's AI systems and practices are evaluated within its responsible AI framework.
CA-P-000022 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Microsoft · Microsoft Responsible AI Standard
This provision articulates a design standard applicable to Microsoft's AI systems development. It establishes privacy protection and security resilience as specified objectives within the responsible AI framework, creating an operational expectation for system architecture and implementation.
CA-P-002086 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
Microsoft · Microsoft Responsible AI Standard
The provision creates an operational framework requiring privacy and security to be embedded in AI system development and deployment rather than added post-implementation. This establishes a baseline requirement for how Microsoft's AI products must be architected and maintained.
CA-P-000168 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Microsoft · Responsible AI
The provision frames privacy and security as architectural requirements rather than optional features, establishing institutional expectations that data-handling practices in AI systems must account for both operational necessity and protective measures. This sets a standard for how organizations should structure AI systems that rely on data access.
CA-P-002073 First tracked Apr 4, 2026 Last seen May 11, 2026 Compare across platforms →
medium Ai automated
Google · Google AI Principles
The clause operationalizes Google's commitment to embed privacy protections at the architecture level rather than as add-on features, requiring that data collection controls and user notification be integrated into product design and governance practices.
CA-P-001911 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Ai automated
Google · Google AI Principles
This provision operationalizes Google's commitment to embed privacy, security, and IP protection into AI development processes. It establishes an institutional standard for how Google's AI systems are designed and deployed rather than as post-deployment modifications.
CA-P-000142 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Ticketmaster · Ticketmaster Privacy Policy
The clause establishes a default profiling and personalization system that operates automatically upon service use, enabling the platform to segment users based on behavioral and preference data for targeted messaging without requiring explicit opt-in authorization for the profiling mechanism itself.
CA-P-001484 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Ai automated
Google · Google AI Principles
This clause establishes Google's operational policy framework for AI development screening and go/no-go decision criteria. The provision conditions deployment authorization on risk-benefit assessment and safety constraint implementation, creating internal governance requirements for project approval.
CA-P-000141 First tracked Apr 3, 2026 Last seen Apr 9, 2026 Compare across platforms →
Hinge · Hinge Terms of Service
This clause establishes Hinge's authority to control which external systems and applications can access or process data and content within its platform. It creates a gating mechanism requiring affirmative authorization before third-party tools, particularly AI/ML systems, can interface with the service.
CA-P-001232 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
OpenAI · Usage Policies
The provision establishes an operational boundary around permissible use of OpenAI's AI systems by restricting conduct that could impair legitimate oversight mechanisms. This serves to preserve the institutional capacity for monitoring and correction of advanced models.
CA-P-001973 First tracked Apr 4, 2026 Last seen May 11, 2026 Compare across platforms →
Luma AI · Luma AI Terms of Service
The clause establishes a contractual restriction on how output generated through the service may be applied, specifically prohibiting its use in developing competing or derivative AI systems. This prevents customers from leveraging Luma AI's outputs as training data or components for their own machine learning development.
CA-P-007076 First tracked May 8, 2026 Last seen May 8, 2026 Compare across platforms →
medium Ai automated
Microsoft · Microsoft Responsible AI Standard
The provision establishes a baseline operational commitment for AI system design and testing prior to deployment. This frames reliability and safety as design objectives integrated into Microsoft's AI development processes rather than post-deployment performance guarantees.
CA-P-002090 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Ai automated
X · X Rules and Policies
This provision clarifies X's operational capacity to implement testing and experimentation protocols necessary for platform development and service optimization. It establishes the institutional framework under which X may deploy experimental features or modifications.
CA-P-000286 First tracked Apr 3, 2026 Last seen May 11, 2026 Compare across platforms →
Microsoft · Responsible AI
This provision establishes Microsoft's internal governance framework for AI development and signals institutional commitment to principle-based AI system design. The public disclosure of the standard contributes to industry transparency regarding responsible AI development practices.
CA-P-002070 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
OpenAI · OpenAI Usage Policies
This provision establishes a procedural requirement that OpenAI customers implement human oversight mechanisms when deploying automated decision systems in high-impact contexts. The restriction functions as a use-case limitation in OpenAI's service agreement, requiring customers to maintain human review as a condition of service use in these specified domains.
CA-P-010653 First tracked May 11, 2026 Last seen May 11, 2026 Compare across platforms →
Google · Google AI Principles
This provision operationalizes Google's stated commitment to responsible AI development by allocating resources toward safety research and establishing information sharing with external stakeholders. The institutional significance lies in Google's undertaking to contribute to industry-wide safety standards rather than operating in isolation.
CA-P-000144 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Microsoft · Microsoft Responsible AI Standard
Sector-specific governance references establish the operational scope and compliance baseline for how Microsoft's AI systems operate across different industries or regulatory domains. This framing allows the standard to accommodate varying requirements across healthcare, financial services, government, and other regulated sectors without requiring separate documentation.
CA-P-000172 First tracked Apr 3, 2026 Last seen Apr 3, 2026 Compare across platforms →
medium Ai automated
Google · Google AI Principles
The provision establishes Google's institutional approach to AI governance and serves as a statement of organizational commitments regarding AI system development. This framing sets expectations for how Google represents its AI decision-making criteria internally and externally.
CA-P-000140 First tracked Apr 3, 2026 Last seen Apr 3, 2026 Compare across platforms →
Microsoft · Microsoft Responsible AI Standard
This provision frames the institutional commitments that govern Microsoft's AI governance framework. The stated principles establish the operational reference points for how Microsoft describes its approach to system design, data handling, and disclosure practices across its AI products and services.
CA-P-000165 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Microsoft · Microsoft Responsible AI Standard
This provision functions as a declarative framework establishing Microsoft's governance position on AI ethics. The articulation of these principles provides a baseline against which Microsoft's AI practices and product development are positioned, though the clause does not specify enforcement mechanisms, measurement standards, or remedies for non-compliance with stated principles.
CA-P-002085 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
Uber · Uber Privacy Notice
The collection of real-time driving performance metrics enables Uber to monitor operational compliance, safety performance, and service quality across its driver and delivery network. This data collection apparatus supports the platform's ability to assess driver conduct and maintain service standards.
CA-P-002025 First tracked Apr 4, 2026 Last seen May 11, 2026 Compare across platforms →

Compliance Governance Intelligence

Monitor specific governance provisions across platforms.

Compliance includes provision-level monitoring, regulatory mapping, and audit-ready analysis.

Start free Start Compliance free trial