Privacy and Security

Learn about the robust privacy and security measures implemented in Cetegra AI Chat to safeguard your data.

Privacy and Security Overview

At Cetegra AI, we prioritize user privacy and data security, implementing comprehensive measures to ensure your interactions are secure and confidential.

Data Privacy Principles

  • Ownership: You retain full ownership of your data, which includes prompts, completions, and training data.
  • Data Isolation: Your data is isolated and is inaccessible to other customers, OpenAI, Microsoft, or any third parties.
  • No Training Utilization: Data is not utilized to refine or improve OpenAI models or other services.

Security Measures

Your data is safeguarded via Azure OpenAI’s secure framework:

  • Azure Hosting in Your Tenant: Services are hosted in your Azure subscription/tenant, within Microsoft’s secure environment.
  • Independent Operation: The service functions independently from OpenAI-operated services.
  • Exclusive Access to Fine-tuned Models: Any models you fine-tune are exclusive for your use.

For privacy or security concerns, our support team is available to assist.

Tracking and Storing Information

In addition to Azure’s measures, Cetegra AI Chat maintains its own security protocols:

  • Message Tracking: We monitor message volume per assistant and the number of tokens per message to generate insightful reports and manage usage limits.
  • User Access Management: User access to assistants is tracked for efficient management.
  • Chat History: With enabled search history, conversations are stored in a protected database.
  • Document Storage: Uploaded documents are stored in Azure Storage Blobs, accessible only to authorized users. Ensure appropriate user access levels to prevent unauthorized access.

Key Elements from Azure OpenAI Privacy Policy

Data Processing and Storage

  • Data Isolation: Prompts, completions, embeddings, and training data:
    • Are not accessible to other customers.
    • Are not used to improve models.
    • Remain inaccessible to OpenAI.
  • Strict Data Use: Data is used solely to provide the service and to monitor compliance with product terms.

Hosting and Interaction

  • Secure Azure Hosting: Managed independently from other OpenAI services like ChatGPT.

Security Measures

  • Transport Encryption: Data in transit is secured using TLS 1.2 or higher.
  • Exclusive Data Access: Logical data isolation ensures exclusive access to allocated resources.

Data Privacy Guarantees

  • Exclusive Fine-tuned Models: These can be deleted anytime by you.
  • Augmented Prompts: Data remains in the designated source location.

Data Storage Features

  • Double Encryption Option: Data can be encrypted with AES-256 and optionally with a customer-managed key.
  • Data Deletion: Users can delete stored data at their convenience.

Abuse Monitoring

  • Content Filtering: Prompts and completions are evaluated for harmful content in real-time.
  • Opt-out Features: Managed customers may apply to opt-out of monitoring under certain conditions.

User Responsibilities

Users are encouraged to maintain security by using strong passwords, enabling multi-factor authentication, and ensuring their devices are secure.

Incident Response

In the event of a security incident, users will be promptly notified, and steps will be taken to address the issue.

Learn More

For more on Microsoft’s data protection and compliance, visit Data, privacy, and security for Azure OpenAI Service.

By adhering to these privacy and security protocols, Cetegra AI Chat guarantees the protection and confidentiality of your data.

← Groups
Discovery integration →