Hugging Face requires model publishers to include structured metadata in their model cards, covering license type, language, and task type, and recommends additional fields for training data, evaluation results, and limitations.
This analysis describes what Hugging Face'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 metadata fields establish the baseline disclosure a model publisher must provide, which directly affects how users understand what they are permitted to do with a model and what risks it may carry.
Interpretive note: The document describes these as platform-level publishing guidelines rather than enforceable legal terms, so the binding nature of compliance with specific fields depends on Hugging Face's separate terms of service.
This general introductory provision was removed as the guidelines shifted toward specific field requirements rather than general mandatory documentation principles.
View full change record →The mandatory metadata fields, including license declarations and intended use sections, determine what legal permissions and risk disclosures attach to each model a user downloads or integrates from the Hub.
How other platforms handle this
You agree not to engage in any of the following prohibited activities: (i) copying, distributing, or disclosing any part of the Service in any medium, including without limitation by any automated or non-automated 'scraping'; (ii) using any automated system, including without limitation 'robots,' 's...
When you use Microsoft services, you must comply with Microsoft's Code of Conduct. Prohibited conduct includes using the services to do anything illegal, transmitting content that is harmful, threatening, abusive, harassing, tortious, defamatory, vulgar, obscene, or otherwise objectionable. Microsof...
You are solely responsible for the content that you post, upload, or otherwise make available through the Services. Udemy may, in its sole discretion, remove or disable access to any content that violates these Terms or that Udemy determines, in its sole discretion, is otherwise objectionable.
Monitoring
Hugging Face has changed this document before.
Receive same-day alerts, structured change summaries, and monitoring for up to 25 platforms.
"Model cards are a form of documentation that provides information about a model's intended uses, how it was trained, its evaluation results, and other important details. Model cards are important for providing transparency about models and for helping users understand how to use them responsibly.— Excerpt from Hugging Face's Hugging Face Model Card Guidelines
(1) REGULATORY LANDSCAPE: Mandatory metadata fields engage EU AI Act technical documentation requirements for AI systems, particularly for high-risk classifications. The FTC has issued guidance on AI transparency that aligns with disclosure of intended use and limitations. GDPR transparency obligations may apply where training data involves personal data. (2) GOVERNANCE EXPOSURE: Medium. Incomplete or inaccurate mandatory metadata, particularly license declarations and intended use fields, may create intellectual property or contractual compliance exposure for organizations integrating Hub-hosted models into commercial products. (3) JURISDICTION FLAGS: EU organizations face heightened exposure under the EU AI Act where model cards serve as part of required technical documentation for high-risk AI systems. California-based organizations should assess whether model card disclosures satisfy state-level AI transparency obligations. (4) CONTRACT AND VENDOR IMPLICATIONS: Procurement teams integrating models should treat model card license fields as binding license terms and verify that the license type disclosed is compatible with their intended commercial use. Discrepancies between model card disclosures and actual model capabilities may create vendor assessment triggers. (5) COMPLIANCE CONSIDERATIONS: Compliance teams should establish a review process to verify that model cards for integrated models contain complete mandatory metadata, that license terms are contractually compatible with downstream use, and that intended use disclosures align with the organization's planned deployment context.
Full compliance analysis
Regulatory citations, enforcement risk, and due diligence action items.
Free: track 1 platform + weekly digest. Monitor: 25 platforms + same-day alerts. No credit card required.
Compliance Governance Intelligence
Need to monitor specific governance provisions?
Compliance includes provision-level monitoring, governance timelines, regulatory mapping, and audit-ready analysis.
Built from archived source documents, structured governance mappings, and historical version tracking.
The metadata fields establish the baseline disclosure a model publisher must provide, which directly affects how users understand what they are permitted to do with a model and what risks it may carry.
The mandatory metadata fields, including license declarations and intended use sections, determine what legal permissions and risk disclosures attach to each model a user downloads or integrates from the Hub.
No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Hugging Face.