LLM hosted locally
Your data remains secure when using our local model, as it never leaves our system.
Our Large Language Model (LLM) is stateless, meaning it does not learn from, store, or reuse your data after responding to your query.
The system does not train any model on your data. Instead, it sends your prompt and any associated files to the LLM to generate responses.
When integrating data sources such as SharePoint, Google Drive, or network folders, the system creates embeddings—vector representations of the content—which are securely stored in a dedicated customer Vector Database instance. These embeddings are used solely for content search purposes
Data retention (For all types of models)
Synced content is retained indefinitely until explicitly deleted by the user, as no automatic retention policy is applied.
Auditing settings can be configured with a specific retention period under Account Settings > Firewall Settings.
You can chose to annonmyise user info in the auditing.
AGAT Software privacy policy:
GPT hosted in OpenAI
When working with OpenAI API they ensure the privacy of your data (not like working with ChatGPT) as explained here:
https://openai.com/enterprise-privacy/
OpenAI does not train our models on your business data by default.
OpenAI Does not own your inputs and outputs (where allowed by law)
LLM hosted in AWS Bedrock
Amazon Bedrock has a concept of a Model Deployment Account—in each AWS Region where Amazon Bedrock is available, there is one such deployment account per model provider. These accounts are owned and operated by the Amazon Bedrock service team. Model providers don't have any access to those accounts. After delivery of a model from a model provider to AWS, Amazon Bedrock will perform a deep copy of a model provider’s inference and training software into those accounts for deployment. Because the model providers don't have access to those accounts, they don't have access to Amazon Bedrock logs or to customer prompts and completions.
Secure Gen AI Apps - Amazon Bedrock Security and Privacy - AWS