Configuring Grounding Rules in Salesforce

Kalanithi Balasubramanian

Kalanithi Balasubramanian

Updated on April 15, 2024

Welcome to the step-by-step guide on configuring grounding rules in Salesforce AI prompts. Grounding rules act as guardrails, ensuring ethical, content-related, dynamic, and data security aspects of your AI responses.

Configure grounding rules to:

  1. Write better AI prompts.
  2. Accelerate revenue.
  3. Slash grunt work.
  4. Make AI safer, trustworthy, and aligned with human values.

There are two types of grounding rules:

  1. Content
  2. Ethical

Set Content Grounding Rules

Content grounding ensures that AI responses are complete, appropriately toned, and correctly formatted.

Here are some prepackaged content grounding rules to guide your AI:

  • Use data from the provided JSON, linked by 18-digit IDs.
  • Use the following inline CSS for table header ‘th’: style=”background-color: #e1ebf7; color: #16325c; padding: 0.5rem; text-align: left;”
  • Use the following inline CSS for caption: style=”font-size: 1.0rem; font-weight: bold; padding: 0.5rem; color: #5C5C5C;text-align: left;”
  • Address all parts of a multi-part question.
  • Aim for clarity over complexity.
  • Avoid using absolutes, such as “always” or “never”.
  • Avoid using passive voice.
  • Avoid using slang or colloquial terms unless specifically asked.
  • Only answer what has been asked for.
  • Refrain from using first-person pronouns.
  • Ensure all records are shown in the tables.
  • Add captions to each HTML table.
  • Convert all internal CSS to inline CSS for the response.
  • Include border=”1″ in all HTML table tags
  • Prioritize quality over speed in responses.

Set Ethical Grounding Rules

Here are some ethical grounding rules:

  • Do not generate content that can be deemed offensive or inappropriate.
  • Avoid using slang or colloquial terms unless specifically asked.
  • Prioritize safety and ethical considerations in your answers.
  • Avoid using emotive language.
  • Do not use scare tactics.
  • Do not exaggerate.
  • Do not generate content that can be deemed offensive or inappropriate.
  • Do not hallucinate
  • Your response must be based on the information provided in the prompt
  • Use data from the provided JSON, linked by 18-digit IDs
  • Task summary must always include names and titles of people involved
    Write summaries in third person and condense key points.

Remember, these rules are just a starting point. You can customize them according to your specific needs.

Dynamic Grounding Rules

Additionally, here are some more dynamic (sample) grounding rules you can set for better AI responses:

  • When referencing dates and amounts, adhere to the USA locale conventions.
  • Adapt responses to user context.
  • Tailor responses to feedback.
  • Personalize based on preferences.
  • Implicitly adjust to user cues.

Conclusion

Grounding rules are essential training wheels for your AI, aligning it with ethics, safety, and corporate values. Use GPTfy to seamlessly automate grounding rules in your Salesforce+AI prompts.

Ready to configure your Salesforce+AI with grounding rules? Get GPTfy Now: AppExchange Link 🚀

Additional Resources:

Salesforce + AI Demo Videos
Salesforce + AI Use Cases Datasheets, and more