6.3: AI in Research

Using AI in the Research Process

Artificial intelligence (AI), including tools like ChatGPT, refers to computer systems designed to analyze information, recognize patterns, and generate helpful responses based on large amounts of training data. AI can be a powerful partner during the research process because it can help you brainstorm topics, summarize complex ideas, generate search terms, or explain difficult concepts in simpler language. However, AI is not perfect, its information is not always accurate, up-to-date, or complete. Because AI generates responses based on patterns rather than verifying facts, it can occasionally produce errors, outdated information, or confident-sounding answers that may not be true.

For public speaking, the most responsible way to use AI is as a creative partner, not a replacement for real research. AI can help you get started by providing ideas, outlines, or examples, but it should never be your only source. You must still find and cite credible evidence from books, articles, experts, and reputable websites. AI can support your thinking, but your role as the speaker is to evaluate accuracy, confirm facts, and build your message with trustworthy sources that you have checked yourself. Strong speakers use AI to enhance their research, not to avoid it.

Examples of How to Use AI as a Research Partner

Brainstorming Topics: Ask AI to generate a list of possible speech topics based on your interests, major, or assignment guidelines, helping narrow your choices more efficiently.

Creating Search Terms: Use AI to develop keywords and phrases they can use when searching library databases, Google Scholar, or credible websites.

Clarifying Complex Ideas: When encountering a difficult academic article or concept, AI can explain it in simpler language so you can better understand what you are reading (while still citing the original source, not the AI).

Outlining a Speech: AI can help generate a preliminary speech outline, which you can refine, expand, and support with credible research you find on your own.

Evaluating Perspectives: Ask AI to list multiple viewpoints on a topic so you can see the larger conversation before confirming those perspectives with real sources.

Practicing Explanations: You can “teach” your topic to an AI tool by explaining it and asking follow-up questions, helping to identify gaps in understanding.

Generating Examples: AI can provide everyday examples, analogies, or scenarios that you can use to illustrate a point, again, only after you verify accuracy and adjusts the examples for your audience.

Using Quality Prompts in AI for Better Research

What is a Prompt 

A prompt is the question, instruction, or request you give to an AI tool to guide its response. In other words, a prompt is how you tell the AI what you want. The clearer and more detailed your prompt, the more accurate and useful the information the AI can provide.

Levels and Types of Prompts (From General to Specific)

AI prompts can range from very broad to extremely precise. As prompts become more specific, the AI can give more focused, relevant, and accurate responses.

1. General Prompt (Broad, Basic): These provide a simple topic but lack detail.

Example: “Tell me about anxiety.”

2. Guided Prompt (Adds Direction or Focus): These give the AI a general topic plus a direction for the response.

Example: “Explain anxiety in college students.”

3. Detailed Prompt (Adds Purpose and Context): These include the goal of the information and who it is for.

Example: “Explain anxiety in college students for an informative speech and include three major causes.”

4. Highly Specific Prompt (Adds Format, Tone, Length, and Constraints):  These give the AI clear expectations for how the information should appear.

Example:  “Explain anxiety in college students for an informative speech. Include three causes, two statistics from credible sources, and a brief example. Write at a college freshman reading level in one paragraph.”

5. Expert/Custom Prompt (Role-Based Prompting): These ask the AI to take on a specific role or perspective.

Example:  “Act as a speech coach. Create a 3-point outline for a speech on anxiety in college students, including evidence types I should look for and questions my audience may have.”

Steps to Improve Your Prompting

1. Start with Your Purpose:  Know what you’re trying to learn

Are you exploring a new topic?, Gathering examples?, Summarizing complex information?

2. Add Context: Tell the AI who you are and what you need

“I am a college student preparing a 5-minute informative speech…”

3. Specify Format and Length: Tell the AI how to organize the answer

“Write in bullet points,” “Create a 3-point outline,” “Use simple language.”

4. Include Constraints or Requirements: Mention what must be included

Number of examples, Statistics, Definitions, Audience type, Reading level

5. Ask Follow-Up Questions: Treat prompting as a conversation

“Give me more examples”, “Rewrite this at a lower reading level”, “What evidence should I look for to support these points?”

6. Verify Information

Always double-check facts with credible sources, AI is a thinking partner, not a final authority.

7. Refine and Revise: If the answer isn’t quite right, adjust the prompt

“Make this more concise”, “Add a comparison”, “Remove technical terms.”

Example \(\PageIndex{1}\)

Example 1: Using AI as a Partner When Researching for a Speech

Maria is preparing a persuasive speech on why college students should build an emergency savings account. Before diving into databases and articles, she uses AI to help her get started. She asks a highly specific prompt: “Act as a financial educator. Create a 3-point outline on why emergency savings matter for college students, and list keywords I should search in library databases.”
The AI gives her a strong starting structure and useful search terms like “financial resilience,” “emergency fund statistics college students,” and “unexpected expenses higher education.” Maria then uses these terms to find credible sources, verify data, and build a stronger, evidence-based speech. AI helped her brainstorm and organize but the actual research and fact-checking came from her.

Example 2: Using AI as a Partner When Preparing for an Interview

Jamal has an upcoming interview for a part-time campus job and wants to feel confident about answering common questions. He uses AI as a practice tool by giving a clear prompt: “Pretend you are a hiring manager. Ask me five common interview questions for a campus student job, one at a time, and give feedback after each answer.”
AI walks him through realistic questions and provides suggestions on how to clarify his strengths, describe past experiences, and improve his confidence. Jamal still prepares his own stories, examples, and qualifications, but the AI helps him rehearse effectively and identify areas where he needs stronger explanations.

Key Takeaways 

• AI tools can be valuable research partners that help students brainstorm ideas, organize information, and clarify complex concepts.
• Because AI is not always fully accurate, students must confirm facts with credible sources and avoid using AI as a substitute for real research.
• Effective prompting, using clear, specific instructions significantly improves the quality, usefulness, and reliability of AI-generated responses.

Exercises

• Prompt Improvement Activity: Use a basic prompt (e.g., “Explain credit scores”) and revise it into a more specific, detailed prompt, then compare the quality of the AI responses.
• AI + Credible Source Check: Use AI to brainstorm three research ideas for a speech topic, then locate and verify at least one credible source for each idea.
• Role-Based AI Practice: Create a role-based prompt (e.g., “Act as a financial advisor…”) to generate initial guidance, then write a short reflection explaining how AI helped and what information still required independent research.