The DPA explicitly states that Personal Data processed under it will not be used to train Perplexity's large language models.
This analysis describes what Perplexity 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
This provision establishes a categorical prohibition on the use of Customer Personal Data for model training purposes, providing enterprise customers with a contractual basis to assert that their data is excluded from Perplexity's AI development activities.
Business customers using Perplexity services are now governed by an updated DPA with two material operational changes. First, subprocessor disclosures have shifted from a static document attachment (Annex 2) to a live online list maintained at https://trust.perplexity.ai/subprocessors, with notifications of changes coming through in-product notification or email rather than through formal amendment. Second, certain Perplexity services (specifically Embeddings API and Perplexity Search) now operate under product-specific data postures and terms that control over the main DPA, meaning the data handling for those services may differ from the baseline agreement. Business customers should review the Trust Center list regularly and check for in-product notifications regarding subprocessor changes, as the updated terms no longer require static amendment cycles.
View change record →This explicit prohibition directly addresses a key customer concern regarding use of their data for AI model development, representing a significant contractual commitment regarding data use limitations.
View full change record →The agreement states that Personal Data submitted through the Services under this DPA will not be used to train Perplexity's large language models, establishing a contractual restriction on this category of data use.
Cross-platform context
See how other platforms handle LLM Training Prohibition and similar clauses.
Compare across platforms →"For the avoidance of doubt, Personal Data will not be used for training of Perplexity's large language models.Excerpt from Perplexity AI's Perplexity Data Processing Addendum
REGULATORY LANDSCAPE: This prohibition engages GDPR Article 5 (purpose limitation), CCPA Service Provider restrictions, and equivalent US state law provisions requiring that data processed under a service provider relationship not be used for the service …
Enforcement risk, jurisdiction flags, contract triggers, and due diligence action items.
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.
This provision establishes a categorical prohibition on the use of Customer Personal Data for model training purposes, providing enterprise customers with a contractual basis to assert that their data is excluded from Perplexity's AI development activities.
The agreement states that Personal Data submitted through the Services under this DPA will not be used to train Perplexity's large language models, establishing a contractual restriction on this category of data use.
No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Perplexity AI.