> For the complete documentation index, see [llms.txt](https://trust.memori.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://trust.memori.ai/en/faq-and-security/is-data-used-for-training.md).

# Is data used for training?

No. Your data is never used to train artificial intelligence models.

This is probably the most important question a customer can ask an AI technology provider. Memori's answer is clear and documented: **users' data — conversations, Agent content, uploaded files — is never used by Memori to train, modify or improve its own models or those of third parties.**

***

### What happens to your data

Data processed through the platform is used exclusively to:

* Provide the conversational service to the Agents' end users
* Allow the Agent's authors to view and manage their own content
* Generate aggregated usage statistics for the Agent's owner

No data is extracted, aggregated or transferred for research, development or model training purposes.

***

### LLM providers and training

The LLM providers integrated in Memori's platform (OpenAI, Anthropic, Mistral, etc.) do not use the data sent via API to train their models. All integrated providers contractually guarantee that the queries sent by Memori are excluded from training their own models.

| Provider               | Use for training | Source                                                                                         |
| ---------------------- | ---------------- | ---------------------------------------------------------------------------------------------- |
| OpenAI                 | No¹              | [platform.openai.com/docs/guides/your-data](https://platform.openai.com/docs/guides/your-data) |
| Anthropic              | No¹              | [privacy.anthropic.com](https://privacy.anthropic.com)                                         |
| Mistral AI             | No¹              | [help.mistral.ai](https://help.mistral.ai)                                                     |
| Microsoft Azure OpenAI | No               | [learn.microsoft.com](https://learn.microsoft.com)                                             |
| Amazon Bedrock         | No               | [docs.aws.amazon.com/bedrock](https://docs.aws.amazon.com/bedrock)                             |
| Google Vertex          | No¹              | [cloud.google.com/vertex-ai](https://cloud.google.com/vertex-ai)                               |

> ¹ = The policy of non-use for training applies to calls made via API. Direct interactions with the consumer products of the same providers (e.g. ChatGPT, Claude.ai) are subject to separate terms.