Configuring Generative AI Controller

Configure Generative AI Controller providers and capabilities.

Configuration overview

Generative AI Controller enables you to choose between several AI providers or a generic connector.

• Microsoft Azure OpenAI
• OpenAI
• Google AI Studio
• Google Vertex AI
• Aleph Alpha
• IBM watsonx

Configure API credentials

Configure your large language model (LLM) credentials to use third-party integrations with OpenAI, Azure OpenAI, Google Cloud, Aleph Alpha, IBM watsonx, and generic models to control the third-party integration.

Configure a generic LLM connector

Set up an external LLM to use generative AI capabilities to add finer control over prompts and transformer scripts.

Configure an AI service provider for a capability

Set your preferred AI service provider for each capability and tailor generative AI content to suit your business needs.

Enable generative AI capabilities in Virtual Agent Designer

Enable generative AI capabilities in Virtual Agent Designer to build custom topics and topic blocks with content created by AI.

Configure sensitive data handling for generative AI

Set up and configure how personally identifiable information and other sensitive data are removed from generative AI prompts.

Add semantic filtering and error handling for sensitive information

Configure semantic filtering of sensitive topics and redirect users to another Virtual Agent topic if sensitive material is brought up in interactions.

Enable recursive summarization for large inputs

Enable recursive summarization to retain context for large inputs in LLM calls.

Enable Dynamic Translation for generative AI capabilities

Use Dynamic Translation for in-product generative AI capabilities to support users who speak different languages.