Appropriate Use of Artificial Intelligence (AI)

Foundations of AI and Ethics

Elder Bednar’s talk emphasizes the importance of truth, integrity, and using technology in ways that align with eternal principles. Key lessons include:

• AI should enhance, not replace, human intelligence and decision-making.

• We must learn how to discern between truth and deception in the digital age.

• AI should be used as a tool for good, rather than as a means of dishonesty or shortcuts.

Ethical Considerations

• Truthfulness: AI should not be used to fabricate information or mislead others.

• Integrity: AI should support, not substitute, human effort in work and education.

• Accountability: Users must take responsibility for AI-generated content and ensure accuracy.

AI Bias and Misinformation

Just like our brains are built from our experiences and learning over time, AI models are also built from what is called "training". Each AI model is simply a product of the information and data that it has been trained on. Because of this, AI models can easily, and often do, reflect biases from their training data. Users MUST critically evaluate AI-generated content rather than just accepting it blindly without review. In the talk, Elder Bednar gave an excellent example of the appropriate use case for AI: 

“I personally do not use this technology to generate or draft my talks, articles, or content for other projects. This must be my diligent effort, my creative work, and, most importantly, my seeking to be open to inspiration from the Holy Ghost. I do use this technology to critique my work.”

This is good guidance for all of us- do not use AI to draft your work or assignments, rather use it as a partner to critique and refine ideas that you have taken the time to develop on your own. If we accept everything that AI generates for us without checking sources or really learning the skill we are trying to learn, we are in danger of appearing dishonest. We are also in danger of what Elder Bednar said- 

"One of my great concerns is that overreliance on AI technology will cause us to become spiritually slothful and shallow—and to forfeit the blessings made possible through righteous work..... But the divine capacities to create and work belong uniquely to each of us as sons and daughters of God. We undoubtedly can generate and produce fabulous content for a sacrament meeting talk with AI. But the objective is not merely producing or presenting impressive content; rather, it is working and becoming what God intends and yearns for us to become. "

Reflection Activity: AI Ethics Dilemma

Objective: To apply the ethical principles of AI use to a real-world dilemma, encouraging thoughtful decision-making and self-reflection.

Instructions: Imagine you are working on a group project in your new job. Your team is tasked with preparing a report using AI tools to analyze data and generate content. One of your teammates suggests using an AI tool to help generate ideas for the report quickly, but the tool sometimes produces content that is factually incorrect. Your teammate insists that you can "fix" the errors later and that it will save time. You're unsure whether to use the tool as suggested.

Reflect on the following questions and use the principles from this section (truthfulness, integrity, accountability, and the ethical use of AI) to guide your decision:

Truthfulness: Is it ethical to use AI tools that might produce inaccurate or misleading content, even if you plan to correct it later? How does this align with the principle of truthfulness?

Integrity: How can you maintain your integrity in this situation? Should you allow the AI tool to generate content, or should you create the content manually to ensure accuracy.

Accountability: Who is ultimately responsible for the final output of the project—AI or the team members? How would you explain the use of AI to your supervisor or team members, and what kind of accountability should be in place for the AI-generated content?

AI and Human Effort: How can you ensure that AI enhances your team's efforts rather than replacing them? In this case, how could AI be used responsibly to assist the team without compromising the quality or accuracy of the report?

Submission: You will be asked to submit this activity in your self-evaluation this week.

Write a short response (200-300 words) discussing how you would handle this dilemma. Be sure to reference the ethical principles of truthfulness, integrity, and accountability in your response. 

Introduction to Prompt Engineering

What is Prompt Engineering, and Why is It Important?

Prompt engineering is the practice of designing effective prompts to guide AI tools in generating accurate, relevant, and useful responses. AI models, such as chatbots and text generators, do not “think” like humans—they rely on the instructions given to them. The quality of their output depends on how well the input (prompt) is structured.

Why is Prompt Engineering Important?

• Enhances Productivity: Well-structured prompts allow users to get helpful responses faster, reducing the need for multiple revisions.

• Improves Accuracy: AI models can generate misleading or incomplete answers if the prompt is unclear. A well-crafted prompt ensures a more precise response.

• Increases Efficiency: In professional settings, AI can assist with research, summarization, report writing, and idea generation—but only if prompted correctly.

• Supports Ethical AI Use: Clear prompts help prevent AI from generating biased, false, or inappropriate content, reinforcing responsible usage.

By understanding prompt engineering, you can optimize AI interactions to enhance your work, studies, and problem-solving skills while maintaining ethical standards.

Basics of Crafting Effective Prompts

Using AI effectively requires clear and well-structured prompts. A poorly written prompt can lead to vague, unhelpful, or even misleading AI responses.

Key Principles for Crafting Effective Prompts:

1. Be Specific: Clearly state what you need instead of making vague requests. 

❌ Poor Prompt: “Summarize this report.”
✅ Better Prompt: “Summarize this 5-page sales report into three key findings, focusing on customer purchasing behavior.”

• Example in the Workplace: You are a customer service representative and need a summary of customer complaints to present in a team meeting. If your prompt is too general, the AI may give an irrelevant summary. A specific prompt will extract only the most important trends.

2. Provide Context: AI works best when it understands the purpose of your request. Tell the AI what the information will be used for and the audience and it will adjust the response accordingly. 

❌ Poor Prompt: “Write an email to a customer.”
✅ Better Prompt: “Draft a polite email to a customer who requested a refund, explaining the refund policy and offering store credit as an alternative.”

• Example in the Workplace: You work in an online store’s customer support team. Instead of a generic email, the AI can draft one tailored to your company’s policies and the customer’s concern, saving you time.

3. Iterate and Refine: If the AI’s response isn’t quite right, tweak your prompt instead of starting over. 

❌ Poor Prompt: “Suggest business ideas.”
✅ Better Prompt: “Suggest three profitable small business ideas for a university student in Africa with low startup capital.”

• Example in the Workplace: You are an entrepreneur looking for ideas to start a side business. Instead of a long list of random business ideas, a refined prompt will give you options that match your budget and situation.

4. Avoid Ambiguous Language: Use clear wording to avoid misunderstandings.

❌ Poor Prompt: “Give me information on marketing.”
✅ Better Prompt: “Provide three digital marketing strategies for a small online clothing business targeting young customers in Nigeria.”

• Example in the Workplace: You are working in a startup and need digital marketing strategies. If your prompt is too vague, the AI might give broad, irrelevant information. A detailed prompt ensures you get strategies suited to your target market.

Do’s and Don’ts of AI Prompting

Do Use AI For:

• Brainstorming new ideas for work and study projects

• Summarizing research papers or lengthy reports

• Generating structured outlines for essays or presentations

• Language improvement (e.g., checking grammar and clarity in writing)
  
  

Don’t Use AI For:

• Generating false or misleading information

• Creating entire work or school assignments without personal input

• Writing job applications or resumes without reviewing and personalizing the content

• Sensitive or confidential topics without understanding data privacy risks
  
  

AI and Data Privacy

AI systems can collect, store, and analyze user data, which can create privacy concerns. It’s important to be mindful of how AI tools handle information.

Key Privacy Considerations:

• Data Storage: Some AI platforms save user queries, which can be used for model improvement or analysis.

• Confidentiality: Avoid inputting personal, financial, or sensitive company information into AI tools.

• Third-Party Risks: AI tools may share data with other systems or organizations, so always check their privacy policies.
  
  

Prompt Engineering Practice

Let's practice evaluating poor AI prompts and writing better ones using the principles we learned above. Here is an example of a poor prompt and an improved prompt. 

❌ Poor Prompt: “Tell me about AI.”
✅ Improved Prompt: “Explain AI in simple terms, with examples of how it is used in business and education.”

Use this example to create improved prompts for the three poor prompts below. You will also share why each one is a poor prompt and why your improved one is better. You will be asked to report on this in this week's self-evaluation.

1. "Tell me about remote work"

2. "Write a good email to my boss"

3. "Tell me a cool idea for my business"

You have reached the end of this module!