AI Assistant Best Practices

Welcome to the Checkbox AI Assistant Best Practices Guide. The Checkbox AI Assistant is an intelligent chatbot (or bot) designed to answer questions from business users, and redirect them to relevant apps and documents when relevant. 

To harness the full potential of the Checkbox AI Assistant, it's essential to set up the sources and instructions accurately. This guide will walk you through the best practices for preparing app sources, document sources, and setting up the Assistant's instructions to get optimal results. 

 

Table of Contents

Preparing document sources

Preparing app sources

Updating the Assistant Instructions

 

 

Source Preparation

Screenshot 2023-08-25 at 10.38.46 am.png

 

Document Source Best Practices

In general, the AI Assistant will perform better on text based documents (e.g. word and pdf files) instead of image or spreadsheet based files.

✅ Document Sources Work Best When:

Example 1: Structured in Q&A Format

Good Practice:

Q: What is the purpose of a Non-Disclosure Agreement (NDA)?
A: An NDA is used to protect confidential information when sharing it with third parties.

Bad Practice:

Information about NDA: Non-Disclosure Agreements are used for protecting confidential information.

Example 2: Hyperlinks Fully Spelled Out

Good Practice:

For more information, visit our website at: www.example.com

Bad Practice:

For more information, click here.

Example 3: Content in a Single Document

Good Practice:

All the chapters of the manual are included in one PDF document.

Bad Practice:

Chapters are spread across multiple PDF documents, requiring navigation between files.

❌ Document Sources Do Not Work Well When:

Example 1: Image-Dense Formats

Good Practice:

The document contains mostly text with occasional illustrative diagrams.

Bad Practice:

The document is a PowerPoint filled with images and minimal text.

Example 2: Hyperlinks Embedded into Words

Good Practice:

For our privacy policy, visit: www.example.com/privacy

Bad Practice:

For our privacy policy, click here.

 

App Source Best Practices

For each App that you add into sources, include:

  1. 1~2 line summary of the app.

  2. Example questions that a user might ask in which we want to trigger the NDA app (e.g. "I want a NDA", I want a non-disclosure agreement")

  3. Describing WHAT the key topic is, as opposed to HOW it works. General information about the key topic is helpful. A helpful litmus test is to describe the topic as though it were NOT a workflow App.

  4. Scenarios: Include scenarios of when you expect this App to be used (e.g. “I want to share information with a vendor”)

--- 

Example of how to setup a NDA app as a source

  1. This App allows business users to self-serve non-disclosure agreements (NDA).

  2. This App allows business users who ask for a NDA (e.g. they ask "I want a NDA", "I want a non-disclosure agreement", "I want an agreement for sharing confidential information") to automatically self serve this agreement.

  3. A non-disclosure agreement (NDA) is a legally binding contract that establishes a confidential relationship. The party or parties signing the agreement agree that sensitive information they may obtain will not be made available to any others. An NDA may also be referred to as a confidentiality agreement. Non-disclosure agreements are common for businesses entering into negotiations with other businesses. They allow the parties to share sensitive information without fear that it will end up in the hands of competitors. In this case, it may be called a mutual non-disclosure agreement.

NOTE: Similarly to training a new colleague, it is sometimes helpful to repeat important parts of your instructions.

 


 

Assistant Instructions Prompt Best Practices

The Assistant Instructions prompt can be used to change the behaviour of the AI assistant.

Screenshot 2023-08-25 at 10.38.03 am.png

Some helpful tips when editing the assistive prompt provided to the AI assistant include telling it to:

  • Play a particular role/occupation

  • Respond in a particular tone/manner

  • Reinforce that it should not lie, nor give generalised information

  • Use the phrase "Think through your response step by step to arrive at the right answer"

  • (Optional) Include specific phrases that you want the AIA to say with every response like "Does that answer your request?"

EXAMPLE

You are a world class in-house legal professional helping non-legal business users in an in-house legal context. Respond to questions in a helpful, conversational, concise and business-friendly manner. Do not give generalised information. Only answer the question that is asked. Think through your response step by step to arrive at the right answer. After a response is given, always append to the end "Does that answer your request?"

 

NOTE: Similarly to training a new colleague, it is sometimes helpful to repeat important parts of your instructions.

 


 

FAQ

When to use a separate AI Assistant vs combining everything in one "mega" AI Assistant?

Generally speaking, limited scope AI assistants performs better than broader scope. Broader scope usually means lots of (somewhat unrelated) source documents are uploaded. This makes it harder for us to retrieve the relevant "chunks" of documents to provide to the AI Assistant as background.

Can the AI Assistant ask clarifying questions?

Currently, this behaviour is not built into the AI Assistant. However, the "Assistant Instructions" can be adjusted to promote this behaviour. In addition, the source document should give enough information for the Chatbot to know what information is required. The best case would be if the source document explicitly calls out the datapoints required for a decision.

How can we give the bot a personality?

We've seen "personified" AI Assistants used as a way to drive adoption and improve user experience. You can setup "personified" Ai Assistants by adjusting the "Assistant Instructions". Fun examples include: "speak like Yoda", "pretend you're Santa", "make spooky references" during Halloween, or even just incorporate company values into responses.