Compliance & confidentiality
Customer data
Section titled “Customer data”This section explains, at a high level, how your information is handled and kept confidential.
What we do
- Limit access to people who need it to run the service and support you
- Use safeguards to protect information in storage and during transfer
- Keep data only as long as needed for the service and legal obligations
What we do not do
- Sell your content or customer data
- Use your data to train public AI models without your control
What this means for you
- Keep your account access private and review who can access your work
- Share only what is necessary for your project and goals
AI data
Section titled “AI data”Contradic uses large language models (LLMs) to generate answers. The models currently set up are:
- GPT-5
- GPT-5.1
- Mistral Medium
What this implies
- Your request is processed by one of these models to produce the response
- The model only sees the text and context you provide for that conversation
- Different models can lead to slightly different styles or depth of answers
Confidentiality of Expert prompts
Section titled “Confidentiality of Expert prompts”AI Experts are curated guides. Their content is protected to preserve quality and trust.
What we do
- Keep Expert instructions confidential
- Ensure Experts operate within the same data boundaries as standard AI chat
What we do not do
- Expose the internal wording of Expert prompts
What this means for you
- You can use Experts with confidence that their guidance is consistent and protected