In the March 23 session of Pryon’s webinar series, Pryon President Chris Mahl and Product Marketing VP David Tareen introduced the Knowledge Bot — Pryon’s AI-powered “booster engine” for chatbot and service agent performance. Browse a few attendee questions and answers below. Click here to view the March 23 on-demand version.
Q: How much content can Pryon handle?
Chris: From tens to hundreds or thousands of documents. We’re sourcing diverse content from diverse repositories through our connectors and we’re building more connectors all the time into the platform. The term I use is we “fuse” it together for all these models so that the content is clean and dialogue-ready. In terms of capacity, it’s limited only by your cloud provider’s horsepower.
Q: What is the licensing or fee structure?
Chris: We have an annual license fee and a per-application fee because we’re seeing clients activate multiple applications for Pryon such as in HR and in sales. Those both represent very different content sets.
Q: Can you speak questions to Pryon?
David: Yes, you can dictate questions using your voice and Pryon will conduct the search and display the answer. And, as we showed in the demo, you can click on the answer and Pryon will take you to exactly where that answer is located within your knowledge base and within that document.
Q: We use Okta, can your platform leverage our existing authentication system?
Chris: Yes, 100%. You don’t have to change your security model — or your content systems or your communication platforms, we integrate with your existing systems.
Q: Can you quickly show the different sources of content that you work with?
David: We can use a URL to ingest web content, as well as Box, SharePoint, Zendesk, Google Drive, Amazon S3, ServiceNow, and Confluence — and we’re building more connectors every quarter.
Q: Can you refresh content collections and how does the system handle that?
Chris: We know articles can change and get updated on a weekly basis; Pryon has a change facility that identifies and works with your authoring systems to figure out what’s new and then it fuses a new vector model with that change included. And all of this is seamless to the users.