You cannot use Meta's Llama 3 model or any content it generates to train, fine-tune, or improve any AI model that is not itself based on Llama 3.
This analysis describes what Meta'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
This clause restricts a core technical workflow in AI development, prohibiting licensees from using Llama 3 outputs as training data for any competing or alternative large language model, which may affect organizations running multi-model AI development programs.
Interpretive note: The scope of 'output or results' is not precisely defined in the agreement, creating ambiguity about whether transformed, filtered, or indirectly derived outputs are covered by this restriction.
Developers and AI research teams who use Llama 3 outputs in any capacity are prohibited from incorporating those outputs into training pipelines for non-Llama models; this restriction applies regardless of whether the competing model is proprietary or open-source.
How other platforms handle this
You may not use the Services to develop foundation models or other large scale models that compete with Amazon Bedrock or any other AWS Service.
You may not use the Services, including any outputs, to develop, train, fine-tune, or improve any machine learning model or artificial intelligence system that competes with AI21's products or services.
You may not use any content generated by Runway's tools or services to train, fine-tune, or otherwise develop competing AI models or products without Runway's prior written consent.
Monitoring
Meta has changed this document before.
Receive same-day alerts, structured change summaries, and monitoring for up to 10 platforms.
"You will not use the Llama Materials or any output or results of the Llama Materials to improve any other large language model (excluding Meta Llama 3 or derivative works thereof).— Excerpt from Meta's Llama Community License Agreement
(1) REGULATORY LANDSCAPE: This restriction may interact with EU AI Act provisions on training data documentation and copyright compliance, particularly where organizations are required to disclose training data sources. The restriction's enforceability may vary by jurisdiction depending on how courts interpret competitive use restrictions in software licenses. No specific regulatory authority directly governs this provision, but the FTC's interest in AI competition practices is relevant. (2) GOVERNANCE EXPOSURE: High. This provision creates an ongoing monitoring obligation for any organization that uses Llama 3 in a multi-model AI development environment. The scope of 'output or results' is not defined with precision in the agreement, creating interpretive ambiguity about whether indirect or transformed outputs are covered. (3) JURISDICTION FLAGS: EU competition law may limit the enforceability of restrictions that could be characterized as anti-competitive in certain market contexts. US antitrust analysis may also engage if the restriction is applied in ways that foreclose market competition, though this analysis is jurisdiction- and fact-specific. (4) CONTRACT AND VENDOR IMPLICATIONS: Organizations procuring Llama 3 for use in AI research or product development must audit existing data pipelines to ensure Llama 3 outputs are segregated from training datasets used for non-Llama models. Vendor contracts that involve AI training data derived from Llama 3 outputs may require amendment. (5) COMPLIANCE CONSIDERATIONS: Compliance teams should implement technical controls and data governance policies to prevent Llama 3 outputs from entering training pipelines for non-Llama models. Regular audits of AI training data provenance are advisable given the ongoing nature of this obligation.
Full compliance analysis
Regulatory citations, enforcement risk, and due diligence action items.
Free: track 1 platform + weekly digest. Watcher: 10 platforms + same-day alerts. No credit card required.
Professional Governance Intelligence
Need to monitor specific governance provisions?
Professional includes provision-level monitoring, governance timelines, regulatory mapping, and audit-ready analysis.
Built from archived source documents, structured governance mappings, and historical version tracking.
This clause restricts a core technical workflow in AI development, prohibiting licensees from using Llama 3 outputs as training data for any competing or alternative large language model, which may affect organizations running multi-model AI development programs.
Developers and AI research teams who use Llama 3 outputs in any capacity are prohibited from incorporating those outputs into training pipelines for non-Llama models; this restriction applies regardless of whether the competing model is proprietary or open-source.
No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Meta.