AI for Academic Success:
Tools, Ethics, and Best
Practices
1. Definitions and background
2. ChatGPT
3. Creating prompts
4. AI and novice users
5. AI in research databases
6. How to cite AI
7. Ethical use
Agenda
1. Definitions and background
What is AI?
AI: Some definitions
Artificial intelligence (AI) is technology that enables computers and machines to simulate human learning, comprehension, problem solving, decision making, creativity and autonomy.
What is artificial intelligence (Ai)? | ibm. (2024, August 9). https://www.ibm.com/think/topics/artificial-intelligence
What is generative AI?
AI: Some definitions
Generative AI, sometimes called gen AI, is artificial intelligence (AI) that can create original content such as text, images, video, audio or software code in response to a user’s prompt or request. What is generative ai? | ibm. (2024, March 22). https://www.ibm.com/think/topics/generative-ai
AI: Some definitions
What does GPT stand for?
Generative Pre-trained Transformer
• Generative means the model can create or generate new content.
• Pre-trained refers to the model being trained on a large dataset beforehand to learn language patterns and structures before being fine-tuned or used.
• Transformer is a type of deep learning architecture that allows the model to understand context and relationships in data by paying attention to different parts of the input simultaneously.
• GPT models are a family of advanced neural network AI models designed for natural language processing tasks, capable of generating human-like text based on the input they receive. GPT models rely on the transformer architecture.
1. Definitions and backgroundhttps://libguides.hofstra.edu/AIinLibrary
View these videos: https://lib.arizona.edu/tutorials/chatgpttech/#/ https://vimeo.com/1089903579?fl=pl&fe=s h
2. ChatGPT – go to https://my.Hofstra.edu/
Click on the My Apps menu and select the ChatGPT Edu icon to access your account.
Agenda
3. Creating prompts
Creating Prompts – 7 Guidelines
1. Be Clear and Specific - Avoid vague language.
State exactly what you want (e.g., “Summarize this article in 3 bullet points” instead of “Tell me about this article”).
2. Provide Context - Tell the model who it should “be” or what perspective it should take.
Example: “You are an undergraduate biology tutor. Explain photosynthesis at a beginner level.”
3. Define the Format - Specify how you want the answer presented (list, essay, dialogue, code, etc.).
Example: “Give me a table with 3 columns: country, population, and GDP.”
4. Set Constraints - Use word limits, style requirements, or levels of detail.
Example: “Explain blockchain in under 100 words using simple language.”
OpenAI. (2025, September 18). Basic rules for constructing good prompts in ChatGPT [Handout generated by ChatGPT]. ChatGPT. https://chat.openai.com/
Creating Prompts – 7 Guidelines
5. Break Down Complex Tasks - For longer or harder prompts, split into steps.
Example: “First outline the main arguments of this paper. Then suggest 3 counterarguments.”
6. Iterate and Refine - Treat prompting as a conversation. If the first response isn’t right, re-prompt with clarifications.
Example: “Make it less technical and more student-friendly.”
7. Use Examples - Show the model what kind of answer you want by giving examples.
Example: “Here’s a sample research question: ‘How does climate change affect migration?’ Now write 3 more in a similar style.”
OpenAI. (2025, September 18). Basic rules for constructing good prompts in ChatGPT [Handout generated by ChatGPT]. ChatGPT. https://chat.openai.com/
Agenda
4. AI and novice users
AI and Novice Users
Generative AI can be fundamentally destructive for knowledge development in novice users if it undermines the conditions under which people normally construct, test, and internalize knowledge.
Here are some of the main risks:
1. Illusion of Understanding
Novices may confuse exposure to an answer with actual understanding.
Because generative AI produces polished, authoritative-sounding responses, learners might accept them passively without engaging in the deeper struggle that builds conceptual grasp.
OpenAI. (2025, September 17). Generative AI and its impact on novice learners [ChatGPT response]. ChatGPT. https://chat.openai.com/
AI and Novice Users
2. Erosion of Productive Struggle
Knowledge development, especially for novices, depends on trial, error, and persistence.
If AI removes that struggle (by offering instant explanations, summaries, or even solutions), the learner may never practice the cognitive moves—like hypothesis testing, error analysis, or schema building—that enable true expertise.
3. Dependence on External Cognition
Offloading recall, synthesis, or reasoning to AI too early risks “outsourcing thinking.”
Instead of building internal knowledge structures (schemas), the novice leans on external tools, which prevents the transition from surface knowledge to deep knowledge.
OpenAI. (2025, September 17). Generative AI and its impact on novice learners [ChatGPT response]. ChatGPT. https://chat.openai.com/
AI and Novice Users
4. False Epistemic Authority
AI outputs can be wrong, biased, or fabricated, but novices often lack the metacognitive skill to detect errors. This risks building fragile or distorted knowledge structures, which can take longer to “unlearn” than to learn correctly the first time.
5. Shallow Engagement with Sources
Traditional knowledge development pushes learners to seek multiple perspectives, read deeply, and wrestle with primary materials. Generative AI compresses and smooths information into simplified outputs, discouraging the rich engagement with sources that develops critical literacy. OpenAI. (2025, September 17). Generative AI and its impact on novice learners [ChatGPT response]. ChatGPT. https://chat.openai.com/
AI and Novice Users
6. Stunted Metacognition
Novices need to practice evaluating their own understanding (“Do I really get this?”).
If AI constantly supplies seemingly definitive answers, it may blunt their ability to recognize gaps, question authority, and monitor their own learning.
Generative AI is not inherently destructive—it can scaffold novices if carefully integrated (e.g., as a tutor that explains reasoning, poses questions, and encourages reflection).
But without intentional design and guidance, it risks creating fragile, dependent learners who mistake fluency for mastery, and whose knowledge development is short-circuited at the surface level.
OpenAI. (2025, September 17). Generative AI and its impact on novice learners [ChatGPT response]. ChatGPT. https://chat.openai.com/
4. AI and novice users
See https://libguides.hofstra.edu/AIinLibrary/AIP recautions for more information on the safe use of AI in academic research.
References
Bruner, J. S. (1961). The act of discovery. Harvard Educational Review, 31(1), 21–32.
Flavell, J. H. (1979). Metacognition and cognitive monitoring. American Psychologist, 34(10), 906–911.
Freire, P. (1970). Pedagogy of the Oppressed. New York: Continuum.
Piaget, J. (1972). The psychology of the child. New York: Basic Books.
Sweller, J. (1988). Cognitive load during problem solving. Cognitive Science, 12(2), 257–285.
Vygotsky, L. S. (1978). Mind in society. Harvard University Press.
OpenAI. (2025, September 17). Generative AI and its impact on novice learners [ChatGPT response]. ChatGPT. https://chat.openai.com/
5. AI in research databases
See https://libguides.hofstra.edu/AIinLibrary/AIin AcademicDatabases
6. How to cite AI https://libguides.hofstra.edu/AIinLibrary/Ho wToCiteAI
7. Ethical use
https://libguides.hofstra.edu/AIinLibrary/Ethi csInAI
All of this material can be found on this libguides:
https://libguides.hofstra.edu/AIinLibrary