This policy describes how Vijnara designs and operates its AI features responsibly. Our approach is simple: answers should be grounded in your own permitted content, traceable to sources, and always subject to human judgment. It complements our Privacy Policy and Terms of Use.
Answers come from your retrieved, permitted content — not invented.
Retrieval respects each user's role and department access.
Citations and audit logs make every response checkable.
Output supports people; it does not replace their judgment.
Vijnara uses Retrieval-Augmented Generation (RAG). When a user asks a question, the platform retrieves relevant passages from the content they are permitted to see, then a language model composes an answer strictly from that retrieved context. This design keeps answers anchored to your knowledge base rather than to the model's general training data.
Every generated answer is intended to be supported by source references back to the originating documents, along with a confidence indication where available. Citations let users verify claims at the source. Where the platform cannot find sufficient permitted context, it is designed to indicate uncertainty rather than fabricate an answer.
We do not use your content to train foundation models. Customer documents, queries and generated answers are processed to deliver the Service to you and are not used to train general-purpose models for us or any third party. Any model tuning or improvement is limited to the scope and configuration agreed with the Customer.
AI output is assistive. Users remain responsible for reviewing answers and for any decisions taken in reliance on them. For sensitive workflows, we provide controls such as maker-checker review for content ingestion, configurable thresholds, and audit trails so that humans stay meaningfully in control.
AI systems can make mistakes. Generated answers may be incomplete, out of date, or misinterpret a source, and retrieval quality depends on the content connected to the platform. Vijnara is not a substitute for professional advice and should not be the sole basis for legal, financial, medical, safety or other material decisions. Always verify important answers against the cited sources.
You may use Vijnara's AI features to search, summarize and retrieve answers from content your organization is authorized to access, in support of legitimate business purposes and in line with your internal policies and applicable law.
You must not use the AI features to:
AI processing follows the same protections described in our Privacy Policy, including encryption in transit and at rest, role- and department-based access controls, and audit logging. Sub-processors, including model providers, are engaged under contractual confidentiality and security obligations and are not permitted to use your content for their own purposes.
As AI technology and regulation evolve, we may update this policy. When we make material changes, we will update the “Last updated” date above and, where appropriate, provide additional notice.
Questions about how we build and operate AI? Get in touch: