AI Safety,  Privacy,and Security.

Kiiren is built around Data Privacy, AI Safety, and Security

You Own Your Data

With Kiiren, your valuable knowledge and data remains fully in your control.

We use a technique called Retrieval Augmented Generation (RAG) that keeps your expert knowledge out of a Large Language Model's (LLM) weights. This means your information is not accessible by other users or organizations of Kiiren.

Think of it as having one of the smartest humans in the world, in a private room, who is going to take an open-book test. The expert in this case is the LLM, and the books are your knowledge that are supplied just at the time of answering. This knowledge is referenced and then instantly forgotten by the LLM, and is not sharable to other users.

This lets us give you full control over who can see what knowledge. If you want to keep some knowledge private in your organization or for yourself, you can. If you want other knowledge to be available for your paying members, you can.

Data Privacy and Security

A first principals approach to AI data privacy combined with Enterprise-grade Security.

Kiiren was built with data privacy at its very core. We recognize the power of AI and that if it is not approached the right way it can harm society and everyone's trust. We took this opportunity to build a new application from the ground up with a focus on privacy and security, and how that can be applied to your knowledge when used in an AI.
‍
With Kiiren you get:

  • Encrypted data storage to prevent unauthorized access
  • Data is encrypted in transit from your browser or mobile device to Kiiren
  • Granular permissions so you control who can access and what AI generates
  • Enable and disable your AI knowledge instantly to tune what your audience sees
  • Restrict access to your expert knowledge to only your paying members

AI Safety and Transparency

Know how the AI answered a question

Large Language Models can "hallucinate" if they don't fully know an answer; this is why we are using the RAG technique to give the AI the right answers at the right time.
‍
With Kiiren you get:

  • An AI that provides citations, keywords, and references for generated responses
  • Users can view, edit, and refine AI-generated insights
  • AI decision-making is explainable, traceable, and accountable
  • AI Assistants are fully configurable by humans using plain language
  • Kiiren participates in the Regen AI Ethics Steering Committee