Generative AI tools are affected by the information and data they have access to and the purpose(s) they have been optimized for.
Because a multipurpose tool like ChatGPT has been trained on many types of data from the Internet, not just scholarly information sources, you may need to prompt it very specifically to get useful results.
In contrast, by using an AI tool that has been designed to search for and retrieve information from open access research databases, you will avoid having to potentially sift through non-scholarly results (e.g. news websites).
Perplexity, Claude, and Google Gemini automatically offer links to sources that are relevant to the generated output. ChatGPT will offer sources when prompted. However, when sources are provided, they may or may not be scholarly or research-based, so you always need to double check its claims.
Note: In addition to responses to user prompts, these tools provide notes and/or links to source materials. You can ask for specific types of sources, and sometimes the results will align with those source types, but not always.
Perplexity, OpenAI's ChatGPT, Claude.ai, Gemini, and others offer deep research, Search, or extended "thinking" options. Generally, employing the deep research or extended thinking options can allow the model to review more sources before providing responses.
Remember: Unless the providers of these tools have brokered some type of publisher agreement, the AI chatbots do not pull from information sources that are behind paywalls, like many academic journals are. This is less of an issue with openly published research in the sciences, health, and other areas where federally-funded grants sponsored the research.
Similar to doing a broad web search for background information about a topic, you can use a GenAI chatbot to ask a question about what is known about a topic in your field of interest. You can then ask what is still not fully understood about the same topic.
After identifying a potential research gap, you can continue working with a multipurpose genAI chatbot to brainstorm possible narrowed topics or specific research questions. This example explores the topic idea of supporting the academic success of neurodivergent students further.
Evaluating the Use of Artificial Intelligence (AI)-Powered Apps to Provide Real-Time Feedback on Social Interactions for Students with Social Communication Challenges
This research could examine the feasibility and effectiveness of using AI-powered apps that analyze social interactions (e.g., through video or audio input) and provide real-time feedback on social cues such as body language, tone of voice, and proxemics. It could explore the types of feedback that are most helpful and how these apps can be personalized to meet individual needs.
When you have a research question that you believe is sufficiently narrowed, you can ask a genAI chatbot about recommended databases to search for relevant academic resources to support your literature review.
I plan to do empirical research on the following topic: What types of real-time feedback provided by an AI-powered app (e.g., visual cues, auditory prompts, text-based explanations) are perceived as most helpful and least disruptive by students with autism spectrum disorder during naturalistic social interactions (e.g., classroom discussions, lunch breaks)?
Please recommend 4-6 academic, scholarly research databases I should search for relevant research literature and prioritize the list beginning with the most recommended database for my search.
In the ongoing Gemini 1.5 example about supporting neurodivergent students, Gemini added information about possible keywords and phrases to use when it recommended databases to search. The result to Prompt 5 can be built upon as shown here.
What types of real-time feedback provided by an AI-powered app (e.g., visual cues, auditory prompts, text-based explanations) are perceived as most helpful and least disruptive by students with autism spectrum disorder during naturalistic social interactions (e.g., classroom discussions, lunch breaks)?
| Prompt Element | Example Prompt |
|---|---|
| State your goal | I am presenting to a group of college students about ethical issues related to artificial intelligence. I need help identifying and organizing ideas. |
| Provide task instructions and form of output | Identify the most important ethical issues related to AI. Create a list of 8-10 key points to make, with several bullets for each point explaining its importance. Cross-reference your recommendations with classical principles of ethics and draw connections between those principles and emerging AI technologies. Suggest visuals for the presentation. |
| Give any constraints, warnings and verification requirements | Use the points in my attached paper. Provide links to all sources. Draw material from respected academic publications only. |
| Provide background and context (include as much as you can) | For context: I have been studying AI ethics and have written the attached paper. I now need to explain these issues in a 15-minute oral presentation. The students have only general knowledge about AI and most have not studied ethical matters. |
| Optional: Specify the role the AI should play | In providing your list, consider yourself to be the presenter, who is a college-aged student very close in age to the audience. |
| Optional: Provide examples for the AI to follow | I have created a start on the presentation and have attached that to this prompt. Please provide a critique of my ideas so far. |
* Based on a model developed by AI engineer Ben Hylak and endorsed by OpenAI President Greg Brockman. Table reproduced from "Skill-up on prompts, " Elon University, 2025 Student Guide to AI.