Get the research letter
Companies change their terms quietly. We read every version and catch what actually changed. One email a week on the changes that matter and what they mean.
The document discloses that Mistral Medium 3.5 includes dedicated API endpoints for content moderation (/v1/moderations) and chat-specific moderation (/v1/chat/moderations), available as callable features.
This analysis describes what Mistral AI's agreement states, permits, or reserves. It does not constitute a legal determination about enforceability. Regulatory applicability and practical outcomes may vary by jurisdiction, enforcement context, and individual circumstances. Read our methodology
The availability of moderation endpoints as a disclosed feature indicates that downstream deployers may use the model to evaluate content for policy compliance; the governance implications depend on how these endpoints are used within downstream platforms, particularly in contexts governed by the EU Digital Services Act or platform-specific content moderation obligations.
Interpretive note: The model card discloses the existence of moderation endpoints but does not describe the moderation criteria, accuracy benchmarks, or data handling terms applicable to content submitted through these endpoints.
Under this disclosure, the API authorizes submission of user-generated or system-generated content to moderation endpoints; downstream platforms using these endpoints may route user content through Mistral's moderation infrastructure as part of their content review workflows.
Cross-platform context
See how other platforms handle Moderation and Content Filtering Endpoints and similar clauses.
Compare across platforms →Monitoring
Mistral AI has changed this document before.
Receive same-day alerts, structured change summaries, and monitoring for up to 20 platforms.
"Moderations /v1/moderations Chat Moderations /v1/chat/moderationsExcerpt from Mistral AI's Mistral Medium 3.5 Model Card
(1) REGULATORY LANDSCAPE: The EU Digital Services Act (DSA) imposes content moderation obligations on online platforms; downstream operators using Mistral's moderation endpoints to fulfill DSA obligations should assess whether reliance on a third-party AI moderation API satisfies DSA accountability and auditability requirements. GDPR applies to personal data processed through moderation endpoints. (2) GOVERNANCE EXPOSURE: Medium. Moderation endpoints process content that may include personal data, user communications, or sensitive speech; the legal basis for processing and the accuracy of moderation outputs are governance considerations for downstream platform operators. (3) JURISDICTION FLAGS: EU/EEA platform operators subject to the DSA face heightened obligations regarding transparency, appeal rights, and auditability of automated content moderation decisions; reliance on a third-party AI API for moderation may require disclosure and human oversight mechanisms. (4) CONTRACT AND VENDOR IMPLICATIONS: Platform operators using moderation endpoints should ensure their DPA with Mistral AI addresses the specific data processing activities involved in moderation, including retention of flagged content, output logging, and subprocessor chains. (5) COMPLIANCE CONSIDERATIONS: Legal teams at downstream platform operators should evaluate whether automated moderation via this API satisfies applicable due process and transparency requirements under the DSA, national platform liability frameworks, or internal acceptable use policy commitments.
Regulatory citations, enforcement risk, and due diligence action items.
Provision-level monitoring, governance timelines, and regulatory mapping built from archived source documents and historical version tracking.
The availability of moderation endpoints as a disclosed feature indicates that downstream deployers may use the model to evaluate content for policy compliance; the governance implications depend on how these endpoints are used within downstream platforms, particularly in contexts governed by the EU Digital Services Act or platform-specific content moderation obligations.
Under this disclosure, the API authorizes submission of user-generated or system-generated content to moderation endpoints; downstream platforms using these endpoints may route user content through Mistral's moderation infrastructure as part of their content review workflows.
No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Mistral AI.