Session

Lessons learned from implementing a law chatbot using Azure Open AI and Semantic Kernel

In this session, I'll show a real-world chat app I implemented using Azure Open AI and Semantic Kernel for a law company. The app uses the Azure OpenAI ChatGPT-4 model to answer questions based on knowledge of Czech legislation stored in a vector database.
I will show that implementing such an application is not easy because users' questions have very different intents. Sometimes, they ask about the current state of the law. In other cases, the question is restricted to a specific year, or the users want to get a summary of a specific bill or paragraph.
I'll explain how to use Semantic Kernel to detect the user intent, transform the question into a query for vector search, and compose the response from the results.
I will also discuss the responsible use of AI and the measures we have taken to ensure the solution's compliance with Czech law.

Tomas Herceg

Microsoft MVP, CEO @ RIGANTI, Author of DotVVM

Prague, Czechia

View Speaker Profile

Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.

Jump to top