Conversational Platform
GPT & Generative AI
LoyJoy GPT automatically answers queries about knowledge, follow-up questions or small talk.
Smart AI Conversations: Elevate Engagement with Precision
Revolutionize Customer Interaction with Multi-Step GPT Integration
-
-
Our advanced GPT integration automatically categorizes inquiries – from knowledge questions to follow-ups and small talk. Incorporate your processes into Process GPT, enabling the AI to precisely identify customer intents, whether service or sales inquiries, and execute the appropriate action fully automated.
-
-
You can ensure the best answer quality by restricting the AI to only answer questions based on your knowledge base. This way, you can ensure that the AI only answers questions that you have approved. You can include any web site content into your knowledge base, or you can upload files. Additionally, you can access your knowledge base via API to automatically update the content.
-
-
The GPT is fully integrated into LoyJoy. You can use it in any experience to answer questions from your customers. Additionally, you can use the AI to choose exactly the right next step for your customer and reroute them to the chat flow that fits their current needs.
-
-
When the AI needs help, seamlessly transition to live chat or asynchronous message, without disrupting the customer experience. The fallback can be tailored to your needs to ensure that it fits your workflow.
-
-
In LoyJoy, you can choose which LLM model you want to use. You can use models directly from OpenAI, models hosted in Europe on Azure, or you can create your own deployment in Azure and use it in LoyJoy. This allows you to use the language model that best fits your needs and complies with your data protection requirements.
How LoyJoy's Knowledge Base Works
LoyJoy's knowledge base is a centralized repository of information created by uploading files manually or via API, or by crawling your website. When a user asks a question, LoyJoy searches the knowledge base for the most relevant text snippets related to the query. These snippets are then provided as context to a large language model (LLM), which generates a factually correct answer based on the given information. This approach, known as 'Retrieval-Augmented Generation' (RAG), ensures that the AI provides accurate and relevant responses to user inquiries, drawing from the approved content within the knowledge base.
Choose your own LLM - GPT-4, Claude 3, or Llama 3
LoyJoy offers a variety of language models, including GPT-4, Claude 3, and Llama 3, with new models being added regularly. Select the model that best suits your preferences and requirements for performance, tone, and data privacy, ensuring an AI-powered conversational experience tailored to your brand and audience.