Web research involves using real-time internet access to find and summarize current information from websites, news articles, blogs, or public data. It's useful for exploring recent events, trends, or general background information that may not be available in academic sources. This approach is best when you need timely or broad context rather than in-depth scholarly analysis.
Deep Research involves uploading trusted, often scholarly, sources like PDFs, journal articles, or datasets. You can upload the sources yourself or use the most current model from your LLM.
Changing the role in a prompt means to ask the AI to take a position or perspective. This technique works by setting context for how the AI should respond. For example:
"Act as a data analyst" may yield structured summaries, charts, or technical language.
"Explain like you're talking to a high school student" will typically produce simpler, clearer language with relatable examples.
"Write as a skeptical reviewer" can generate critical analysis or highlight potential weaknesses.
By specifying a role, you guide the AI's perspective, language style, and level of detail, resulting in responses that are better aligned with your intended output.
Uploading files to an AI system allows you to interact directly with the content for a variety of purposes. For example:
For data files, AI can also perform calculations, visualizations, or code generation, making file uploads a flexible tool for research, teaching, and content development.
When uploading files to an AI system, you can also specify the format in which you'd like the output, making the interaction even more tailored and useful. For example, you can ask the AI to:
Give it structure (table, bullet points, etc...)
Limit by word count
Output specific file types (doc, pdf, csv, etc...)
You’re a [role], generate an [output] for a study on [source] including [structure].
Summarize the main findings on [sources], with a focus on the [research topic]. Highlight common themes, such as [include your themes here], and note any gaps or areas for further exploration.
The CLEAR Framework is a practical guide for prompt engineering that helps users create more effective interactions with AI language models by focusing on five key principles: Concise, Logical, Explicit, Adaptive, and Reflective. It is designed to support information literacy and critical thinking in academic contexts by improving the clarity, relevance, and precision of AI-generated content.
| CLEAR Principle | Summary | Example |
|---|---|---|
| Concise | Prompts should be brief and focused, avoiding unnecessary detail to enhance clarity and precision. | “Explain the significance of photosynthesis in plant biology.” |
| Logical | Prompts should follow a clear structure or sequence to help the AI understand relationships between ideas. | “Describe the steps of the scientific method, starting from forming a hypothesis to drawing conclusions.” |
| Explicit | Prompts must clearly specify the desired output, including content, scope, and format, to avoid vague results. | “List five renewable energy sources and briefly explain how each one generates power.” |
| Adaptive | Prompts should be tailored based on context or revised in response to weak outputs. Be flexible and experiment. | “Analyze how social media use correlates with anxiety levels in teenagers, citing recent studies.” |
| Reflective | Prompting is an iterative process; assess AI responses and refine prompts to improve future outputs. | “Provide time management strategies for first-year university students balancing coursework and part-time jobs.” |
Lo, L. S. (2023). The CLEAR path: A framework for enhancing information literacy through prompt engineering. The Journal of Academic Librarianship, 49(4), 102720. https://doi.org/10.1016/j.acalib.2023.102720
