Tipos de indicaciones

What is a Prompt?

A prompt is a question or instruction that you give to an AI system to get the information or response you need.

In a church setting, you might prompt an AI Chatbot to help you prepare for a Bible study. For example, you could ask, "What are the key themes of the book of Romans?" The better your question, the more helpful the answers will be.

How do you use a prompt?

Open your chatbot (ChatGPT, Gemini or others):

Type in your prompt, like you might type in a search question in Google.
Click the arrow or ‘go’ button.
The chatbot gives a response.

Start a new chat for a new topic

The chatbot builds on what has gone before in your chat.

THIS IS GOOD when you keep asking questions to dig deeper into your topic (see also “Follow on Prompts” later in this chapter).

Prompt 1: Can you explain Neuroscience to me? (wait for answer)

Follow on prompt 2: Can you explain it in 50 words and in a way that an 8-year-old can understand?

Follow on prompt 3: Can you give me a simple metaphor to explain it using a bucket and sand?

THIS IS NOT SO GOOD when you are talking about different topics.

If you are asking a question about Football and the rules of the game, then ask about a recipe for a rice dish, it will answer it with a link to soccer! Which might give some interesting results!

Prompt 1: Explain to me the key rules of Football, because I am a dad coaching my son’s 8-year-old soccer team for the first time.

Start new chat.

Prompt 1: Can you give me a recipe for rice and fish that uses coconut milk?

REMEMBER to start a new chat each time you start a whole new topic.

Simply explain what you want

To get good results from an AI system, you need to know how to write good prompts. Here are some starting principles for writing effective AI prompts:

Be specific about what you want the AI to generate.
Use clear and concise language.
Provide as much detail as possible. Use keywords that are relevant to the task that you want the AI to perform.

The following takes you deeper into prompts. It is not as difficult to understand as it might look at first!

Positive Prompts

A positive prompt tells the AI exactly what you want. It is like giving clear instructions to a program leader who will prepare something for you. They need to know exactly what is expected. Except, in this case, they can read and write everything in world-record time!

Prompt: List Bible verses that talk about God's mercy and grace.

Negative Prompts

Prompt: Give a list of Bible verses about leadership, but do not include verses from the Old Testament.

This prompt directs the AI to give answers from the New Testament only, helping us focus on the life of Jesus and the early church’s teachings on leadership.

Follow on Prompt

To get the answer you need, you may need to keep asking questions, building on what has been answered by the chatbot. You don’t need to be satisfied with the first answer. Guide the Chatbot by asking more questions and give more negative and / or positive prompts. You don’t need to repeat the whole question, as you see in this example below:

Prompt 1: Can you give me a game that helps teach the Parable of the Sower from Matthew 13.

Follow on prompt 2: I am doing it with 8–11-year-olds.

Follow on prompt 3: Can you give me a game using only two chairs and a piece of paper?

Follow on prompt 4: I only have five minutes to play the game.

Follow on prompt 5: Can you give me a version with no materials at all.

Elements of a Prompt

The elements of a prompt include the subject (what you are asking about), context (the setting or background), and instruction (what you want the AI to do). Think of these as key elements in the careful understanding of the Bible material, or the preparation of any form of communication with your ministry partners.

For example, when teaching about Jesus fulfilling Old Testament promises, you might ask:

Subject: Jesus and the Old Testament
Context: Focus on how Jesus is the promised Savior
Instruction: Show examples of prophecies fulfilled in the New Testament

An AI prompt might look like this:

Prompt: Explain how Jesus fulfilled the prophecies in Isaiah 53 and connect them to the New Testament.

This prompt contains a clear subject (prophecies about Jesus), a context (Isaiah 53), and an instruction (make connections with the New Testament).

Zero-Shot Prompt

A zero-shot prompt is when you ask the AI to do something without giving it any examples.

Prompt: Explain the parable of the Good Samaritan.

Here, you are asking the AI to respond without giving it any examples or extra context. It is a little bit like asking a new volunteer in your youth program to run a discussion on a parable they are not familiar with! They might focus on any aspect of the story. They might give a good response based on the correct passage in the Bible, or they might prepare something strange based on another story with the same name!

Zero shot prompts can work, but you will receive better responses with more content in your prompt. The following prompt ideas will help.

Few-Shots Prompt

A few-shots prompt is when you give the AI a few examples to guide it. This is like when we train people by giving them a few examples of how something is done, and then they follow that pattern.

Prompt: Explain the meaning of Jesus’ Parable of the Lost Sheep in Luke 15. For example, the lost sheep represents those who are far from God, and the shepherd represents Jesus seeking the lost.

By giving the AI an example of what you want, you help it understand the type of response you're expecting.

Chain of Thought (COT) Prompt

A Chain of Thought (COT) prompt encourages the AI to explain its reasoning step-by-step.

Prompt: Explain step-by-step how the story of Abraham’s willingness to sacrifice Isaac in Genesis points to Jesus’ sacrifice for our sins. Give me the Bible references and quotes from the NIV version.

Here, you’re asking the AI to break down the theological connections between the Old Testament and the New Testament, step by step. This approach is helpful for deeper understanding, much like walking through the Bible’s storyline to show how it all points to Jesus. The Bible references allow you to check the response with your Bible.

Contextual Prompting

Contextual prompting involves shaping an AI response by providing specific background information or guidelines, ensuring the answer is tailored to a particular context.

In ministry, we adapt our teaching or leadership to reflect the unique cultural, theological, or practical needs of the people we serve. We can also ask AI to respond to us in ways that are relevant and aligned with the values, traditions, or doctrines important to our context.

This method is especially useful in theological, literary, or legal contexts where consistency with a specific set of ideas, texts, or doctrines is crucial. For example, in ministry, contextual prompting ensures that the AI’s answers reflect biblical and theological principles—whether it’s emphasizing God’s grace through Christ or the servant leadership model of Jesus. By clearly framing the context, we can guide the AI to provide responses that are both accurate and meaningful for the intended audience.

Depending on the situation, contextual prompting can be understood in various ways:

Framing Prompts

This method applies a specific interpretative lens, such as theological or philosophical, to guide the AI's responses. For instance, you might ask the AI to interpret biblical passages through a Reformed theological perspective, ensuring that the answers reflect the belief that Jesus is the fulfillment of the Old Testament.

Prompt: Interpret the story of David and Goliath with a focus on God’s sovereignty and the foreshadowing of Christ’s victory over sin.

Source-based Prompts

Here, you specify the use of texts or sources, such as Scripture in the NIV or the writings of early church fathers, to ensure the AI draws its answers from resources that will give you the consistency and clarity you need.

Prompt: Using only the NIV translation, explain how Ephesians 2:8-9 highlights the concept of salvation by grace through faith.

Thematic Prompts

This approach focuses on defining a theme that will shape the AI’s response. For example, you might ask the AI to focus on God’s grace and mercy in Christ, and the AI will then frame its response around that central theme.

Prompt: Explain the life of the Apostle Paul, highlighting the themes of God’s grace and mercy in Paul’s conversion and ministry.

Applying Context

When using contextual prompts, it’s essential to tailor your prompts to specific elements such as culture, age, style, and educational level within your ministry. Just as we adapt our approach when ministering in different environments, crafting prompts that reflect these factors will result in responses that are more relevant, accurate, and helpful. Whether you're leading in a rural African village, an urban Indian church, or a large city in Asia, recognizing the unique characteristics of your audience is crucial.

Here are examples of how each element can shape a prompt:

Cultural Context

Cultural values and leadership styles vary widely across regions. For example, in some communities, servant leadership might be expressed through communal care, while in others, it might emphasize personal humility and sacrifice.

Prompt: In a rural (add your country name) village, how can the church model servant leadership in a way that reflects the cultural importance of community and shared responsibility?

In this prompt, the local culture’s emphasis on communal living is taken into account, ensuring the AI suggests culturally appropriate ways to apply servant leadership.

Age Context

The way we engage with children, youth, or elderly members of a congregation can vary. Younger audiences might prefer interactive methods, while older members might resonate more with traditional approaches.

Prompt: For a youth group of people aged 13–17-years-old in an urban (add country name) church, what are some interactive ways to teach about Jesus’ call to ‘love your neighbor’ from Luke 10:27?

Here, the age of the audience (youth) is the key context, encouraging the AI to suggest interactive or engaging approaches suitable for younger people.

Stylistic Context

Different churches have varying worship styles, from formal liturgy to contemporary and charismatic settings. Understanding the stylistic preferences of your audience ensures that the AI provides relevant ideas or recommendations.

Prompt: In a contemporary urban church with a vibrant music ministry, what are 10 ways we could incorporate the theme of ‘God’s sovereignty’ into our worship?

This prompt accounts for the church’s style—contemporary and musically focused—guiding the AI to suggest ways that align with the worship context.

Educational Level Context

Educational background can influence how you communicate theological ideas. In settings where members have varying levels of Biblical literacy, it’s helpful to adapt the depth of explanation or complexity of concepts.

Prompt: In a rural congregation with new believers, explain the concept of justification by faith from Romans 3:28 in simple terms and in 30 words.

Here, the educational level is considered, encouraging the AI to provide an explanation that’s accessible to those who may not have deep theological training.

Geographic or Socioeconomic Context

The socioeconomic realities of a congregation can also play a significant role. A rural congregation with limited resources may have different needs or priorities compared to a wealthier urban church.

Prompt: In a low-income church community in Asia, what practical ways can we demonstrate God’s love and care for the poor, based on James 2:14-17?

This prompt considers the socioeconomic context, prompting the AI to suggest practical ways to live out biblical teachings on caring for the poor.

You could add to this:

Prompt: What are some surprising ways to do this with no money or resources?

Prompt Frameworks

Prompt frameworks provide a structured approach for crafting clear and effective prompts. Just like preparing a Bible study, where we likely follow a clear pattern or method, a prompt framework helps us organize our thoughts and ensure that our requests to AI are clear and precise.

Here are several key frameworks you can use for constructing effective prompts:

1. S.O.C. Framework (Subject, Objective, Context)

This framework ensures that your prompt is well-rounded by including:

Subject: What is the main topic?
Objective: What do you want the AI to achieve?
Context: In what setting or situation will the response be used?

Example:

Prompt: Explain how the parable of the Good Samaritan shows Jesus’ teaching on loving your neighbor. The response should focus on how this applies in a rural church in India.

Here, the subject is the parable, the objective is to explain Jesus' teaching, and the context is set within an Indian church setting, making the response more relevant.

2. R.A.P. Framework (Role, Action, Parameters)

This framework helps you guide the AI by specifying:

Role: Define the AI’s role, such as a teacher, theologian, or servant leader.
Action: What action should the AI take? Should it explain, list, compare, etc.?
Parameters: Set boundaries or specific instructions for the response.

Example:

Prompt: As a Bible study leader, explain the significance of Jesus’ role as the fulfillment of Old Testament prophecy in Isaiah 53. Focus on the theme of suffering and redemption.

Here, the role is a Bible study leader, the action is to explain, and the parameter is focusing on Isaiah 53 and the theme of suffering and redemption.

3. T.E.L. Framework (Theme, Examples, Lens)

This framework is ideal for thematic discussions:

Theme: What is the central theme or focus (e.g., grace, faith, servant leadership)?
Examples: What examples should AI include?
Lens: Through which lens should the response be viewed (e.g., theological, cultural, practical)?

Example:

Prompt: Discuss the theme of grace and mercy in the life of the Apostle Paul, using his conversion and ministry as examples. View this through the lens of servant leadership in an African church context.

Here, the theme is grace and mercy, the examples are Paul’s conversion and ministry, and the lens is servant leadership in Africa.

4. P.I.E. Framework (Problem, Instruction, Explanation)

This framework is helpful for addressing challenges or solving problems:

Problem: What issue or challenge needs to be addressed?
Instruction: What do you want the AI to do about the problem?
Explanation: Ask the AI to provide an explanation or solution based on the given problem.

Example:

Prompt: A rural church in Central America is struggling to engage its youth. Provide three practical ways to address this issue, emphasizing the role of biblical mentorship and using the examples of Paul and Timothy in the Bible.

Here, the problem is low youth engagement, the instruction is to provide practical solutions, and the explanation centers on mentorship and biblical examples.

5. C.H.A.P. Framework (Context, History, Application, Perspective)

This framework is useful for deeper theological or historical prompts:

Context: What is the immediate background or setting?
History: What historical or biblical background is relevant?
Application: How should this be applied to today’s setting?
Perspective: Which perspective or worldview should be used (e.g., reformed theology)?

Example:

Prompt: Using reformed theology, explain how God’s covenant with Abraham in Genesis 12 foreshadows the coming of Christ. Apply this to a modern-day church setting in India, focusing on God’s plan of redemption.

In this case, the context is Genesis 12, the historical background is the Abrahamic covenant, the application is for a modern church in India, and the perspective is rooted in Reformed theology.

Adding Your Ideas

We would love you to add any thoughts you think needed to be add to this chapter on Prompts and Types of Prompts in the Google Form here: https://forms.gle/3egczKxGz6tqGEDu6