If you have any questions on the topics of data access, privacy, information security or related topics, please email [email protected] or [email protected] and we will get back to your promptly.
Where does our data sit?
Your data sits on AWS servers. Here is a link to additional information regarding data protection and privacy on AWS.
All customer data is processed and stored in the US. We do not currently store data in Europe or in other countries. You own all rights to the data, any GDPR requests can be fulfilled by emailing [email protected].
How do you guarantee that only the approved users have access to a certain project?
We implement industry-standard best practices around data access, including both organisation-level filters (every customer has their own URL, firmname.myjunior.ai) and role-based permissions (matching your internal policies). We maintain row-level permissions at the database level (the most granular degree of control possible) to ensure that only users with the right access controls can even see the data.
All client data is classified as confidential and can only be accessed / edited by the client staff members that are given permission to access a particular project / engagement.
We only allow OAuth / SSO to ensure the highest degree of security on login for your organisations’ users. We do not allow email and password logins.
How is our data protected?
All access to Junior is over HTTPS and TLS 1.3. We implement SSL certificates. Our database vendor (Supabase) encrypts all data at rest using AWS EBS volumes encrypted at rest: https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/EBSEncryption.html.
When we call OpenAI APIs, the API is only available over TLS and therefore requests and responses are encrypted.
What do we own? What do you own?
You retain all commercial rights to the data uploaded to the platform.
In the future we may request from you the right to use your anonymised / ‘scrubbed’ data as inputs into a model to help improve the quality of Junior output. This is not currently the case and we will inform you if it changes. Your use of the tool will not be impacted by a refusal to provide training data to a fine-tuned model.
You use Large Language Models to power your application. How does our data interface with these models?