Before using any AI tool for classwork or assignments please check with your instructor first to determine their policy on using ChatGPT and similar AI tools. Each course may have specific guidelines or restrictions on the use of AI. Ensure that you understand and adhere to your course requirements to avoid any issues.
Types of tools (grounded and not-grounded)
Generative AI tools can be categorized into two types: grounded and not-grounded.
Not-grounded: These tools rely on their training data to produce outputs. They don't connect to live, real-time sources of data (such as the Internet) to acquire new knowledge. Therefore their knowledge is fixed as of their last update.
Grounded: These tools continuously learn and update their data through connections to external data sources such as databases or the internet.. As a result these tools can reference up-to-date statistics and more recent information.
Below is a list of both grounded and not-grounded tools. The tools listed provide links to credible scholarly sources and quality web content, which helps minimize errors such as fabricated citations (known as hallucinations). However, always approach results from AI tools with a critical eye and verify any data sources provided to ensure accuracy.
Grounded |
Not Grounded |
Google Gemini - generative AI chatbot developed by Google | Claude - This is an AI assistant which can answer questions and summarize documents |
Microsoft Copilot - Copilot is an AI chatbot available to all UNF students, faculty and staff. It is integrated with Microsoft 365 apps such as Word, Excel, PowerPoint and Outlook. | ChatGPT - This tool uses machine learning to process and generate responses to user queries. |
Academic Research Assistant Tools
The CLEAR Framework
Leo S. Lo outlines five components which users should employ when crafting search prompts using AI tools. The following concepts come from the article cited below:
Concise: "brevity and clarity in prompts"
Logical: "structured and coherent prompts"
A logically structured prompt could be, “Describe the steps in the scientific method, starting with forming a hypothesis and ending with drawing conclusions"
Explicit: "clear output specifications"
Adaptive: "flexibility and customization in prompts"
Reflective: "continuous evaluation and improvement of prompts"
After receiving AI-generated content on the benefits of a plant-based diet, evaluate the response's accuracy, relevance, and completeness. Use insights from the evaluation to refine future prompts, such as asking for more specific benefits or focusing on certain aspects of a plant-based diet.