13.2 AI-Powered Content Generation & Improvement

EPC Modelers can effortlessly generate new content and enhance existing content using EPC’s AI capabilities. Whether creating end-to-end business processes, building a library of risks and controls, or exploring improvement possibilities, our AI can generate or enhance documentation and objects. Users also have the flexibility to manually refine the content during the creation process.

In the following sub-sections, you’ll find detailed guides on:

• Generating EPC Objects
• Generating and Improving Text in Rich Text Fields
• Process Improvement

Configuring AI in EPC

Before using AI features, the AI engine must be configured in EPC by a System Administrator.

EPC supports integration with Large Language Model (LLM) providers such as OpenAI or other compatible API-based services.

1. Obtain an API Key
   • Log in to your LLM provider account (e.g., OpenAI or another supported provider)
   • Navigate to the API or developer section
   • Access the API Keys section
   • Generate a new API key
   • Copy and securely store the key
2. Configure AI Settings in EPC
   In Advanced System Settings, configure the following:
   • AI_FEATURES_IN_UI: Define which AI features are available in the interface. Available values:
     • (empty) → No AI features displayed
     • generate → Enable object and text generation
     • improve → Enable object and text improvement
     • generate, improve → Enable both
   • LLM_API_KEY: Paste your LLM provider API key in this field.
   • LLM_CHAT_MODEL: Specifies the model used for AI-generated responses. Examples: gpt-3.5, gpt-4, provider-specific model names.
   • LLM_CHAT_MAX_TOKENS: Defines the maximum number of tokens generated per request. We recommend a minimum of 2000 tokens for optimal results.

!Limited AI token budget will be provided for free each month. Contact sales@interfacing.com to increase your allocation.

3. The AI content generation feature will be ready for use within your EPC system.

Advanced Configuration (Optional)

These advanced settings allow you to fine-tune AI behavior or connect EPC to custom or enterprise LLM environments. They are not required for standard setup.

• LLM_CHAT_TEMPERATURE: Controls the randomness and creativity of AI-generated responses.
• LLM_CHAT_TIMEOUT: Defines the maximum time (in seconds) the system waits for a response from the LLM provider.
• LLM_CHAT_URL: Defines the API endpoint used for content generation and improvement requests.
• LLM_EMBEDDINGS_URL: Defines the API endpoint used for the AI Assistant.
  • The last two settings are primarily used when connecting to a private LLM deployment instead of the default provider endpoint.