Embedded Files
Newsflash: It's NOT Intelligent!
Newsflash: It's NOT Intelligent!
Introduction
Introduction
You’ve probably used AI to write an essay, make art, or even suggest meals, but is it actually smart?
Not really. AI might look smart, but it doesn’t think like a human. Large Language Models (LLMs) simply predict what word comes next based on patterns they’ve seen in massive amounts of text. They don’t truly understand the content they are generating.
This is why prompt engineering is so important. The better your prompt, the better the AI’s output. Let's learn more!
What is Generative AI?
What is Generative AI?
Generative AI is a type of artificial intelligence that creates new content (text, images, code, music, etc.) rather than just analyzing or predicting from existing data.
Examples:
ChatGPT, DALL·E, MidJourney, GitHub Copilot.
This is unlike traditional AI, which might predict tomorrow’s weather or your next click. Generative AI can produce original outputs based on patterns it learned from massive datasets.
Image generated by Microsoft Copilot using AI image generation tool, December 15, 2025.
Large Language Models (LLMs)
Large Language Models (LLMs)
LLMs are AI models trained on massive amounts of text to understand and generate human-like language.
They don’t “think”, they predict the next word in a sequence based on what they’ve learned.
Capabilities:
Capabilities:
Writing essays, summaries, or code
Translating languages
Answering questions conversationally
Limitations:
Limitations:
Can produce incorrect or misleading information
Doesn't have a true understanding of what it is generating
Sensitive to how you ask questions (the prompt)
Prompt Engineering 101
Prompt Engineering 101
A prompt is the instruction you give to an AI.
The better your prompt, the more accurate, relevant, and useful the AI’s output.
Small changes in wording, context, or constraints can drastically change results.
1. Be Specific
1. Be Specific
General prompt → vague output
Specific prompt → focused, useful output
Example:
Weak: “Explain AI.”
Strong: “Explain AI to a first-year college student in simple language, using one real-world analogy, in under 150 words.”
2. Provide Context
2. Provide Context
Tell the AI who it’s talking to or what it’s about.
Example:
“Explain cloud computing in 3 sentences for a beginner who just learned how to use a computer.”
3. Set a Personal or Role
3. Set a Personal or Role
AI can adopt roles for more targeted responses: tutor, chef, marketer, scientist, etc.
Example:
"You are a computer science professor at MIT. Explain how computers work in layman’s terms so a high school student can understand. "
4. Specify Output Format
4. Specify Output Format
Ask for a particular style, length, or format:
Short summary
Step-by-step instructions
Table or bullet points
Professional email tone
Example:
Weak: “Tell me about cybersecurity.”
Strong: “Explain basic cybersecurity concepts in a table with three columns: Concept, What It Means, and Why It Matters for Everyday Computer Users. Keep it under 10 rows.”
5. Use Iteration and Follow-Up Prompts
5. Use Iteration and Follow-Up Prompts
Don’t expect perfection on the first try.
You can refine output by asking follow-up questions:
“Rewrite this for a younger audience.”
“Make it 50% shorter.”
“Add one real-world example.”
You can even ask it to ask you questions that would improve the prompt!
Example:
“What questions would you need to ask me to give a better explanation for a college freshman?”
6. Combine Techniques
6. Combine Techniques
Mix specificity, context, persona, format, and iteration for the strongest prompts.
Example:
Weak: “Give me a cheap meal idea.”
Strong: “You are a college nutrition coach. Suggest a quick, healthy meal under $10 that a student can cook in 15 minutes. Include ingredients, step-by-step instructions, and explain why it’s budget-friendly. Ask follow-up questions if you need more info about dietary preferences.”
Source
Source
Generative AI (ChatGPT, OpenAI) was used to assist with drafting and refining instructional examples related to prompt engineering. The instructor reviewed, edited, and verified all content for accuracy and appropriateness.
OpenAI. (2025). ChatGPT (GPT-5.2) [Large language model]. https://chat.openai.com/
Page updated
Google Sites
Report abuse