Skillsoft CAISY (the Conversation AI Simulator) Security Standards

Scott Elvin
Scott Elvin
CAPTAIN
Third Anniversary 25 Likes 10 Comments 5 Awesomes
edited December 2023 in Percipio

As with any new technology, there are always questions around how safe it is. We're here to share that we are building CAISY (the Conversation AI Simulator) using Microsoft Azure OpenAI to maintain high security standards.

Skillsoft's CAISYTM (Conversation AI Simulator) is developed and hosted on Microsoft Azure OpenAI and does not use the public ChatGPT service. 

  • Overview: Azure OpenAI is a service that uses the same models as OpenAI, but with the security capabilities of Microsoft Azure.
  • Encryption: Azure OpenAI automatically encrypts your data. The encryption protects your data and helps you meet your organizational security and compliance commitments.
  • Data Security: Azure OpenAI offers private networking, regional availability, and responsible AI content filtering.
  • Azure OpenAI Limited Access Review: Modified Abuse Monitoring - Microsoft Azure does not log or review any OpenAI conversation (inputs/outputs). Logging of conversations is based on opt-in/out or explicit sharing when reporting feedback within the simulator.

Below you will find a document that describes CAISY and includes this information and also describes the guardrails we have put in place to create a high-quality experience for our users (minimize hallucinations, inappropriate or inaccurate responses from AI). Please direct any questions you might have toward your account team who would be happy to follow up on your behalf.

Categories