Making Generative AI Work for You

Making Generative AI work for You!

Matt EdelenNMHED-AE Program Specialist matt.edelen@hed.nm.gov

How do you currently use AI?

If you are using AI tools in your classroom, what tools or applications are you currently using and how do you use them in the classroom or with your students?

Benefits of Generative AI

Personalized Learning

Time-Saving

Data-Driven Insights

Generative AI can tailor content and exercises to the learning style and needs of each adult learner.

By automating routine task like grading and lesson planning, you can free up time for meaningful student interactions.

AI helps analyze student performance data to identify areas for improvement and optimization of teaching methods.

Risk Mitigation: Regular audits and risk assessments help to safegaurd against data breaches or misuse.

Best Practices for Security: Encrypt data, limit data access, and anonymize student information when using AI tools.

Data Privacy Concerns: AI tools in education need to protect sensitive personal data and comply with privacy regualtions.

Generative AI and FERPA- Educators must ensure that any AI tools used do not share or mishandle student data in ways that violate FERPA. You should limit AI's access to or anonymize student data and obtain consent when necessary to ensure compliance with FERPA.

Safety and Data Security

Prompt Engineering

• Designing Precise Inputs: The process of creating well-crafted prompts that lead to the desired AI output
• Iterative Refinement: Effective prompt engineering requires testing and refining to improve AI responses.
• Balancing Clarity and Flexibility: Prompts must be clear, but also allow AI enough flexibility to generate creative and nuanced responses.

Video created using VEED.io

Principles of Effective Prompts

• Specificity: Effective prompts provide clear, detailed instructions to guide AI toward the desired outcome.
• Conciseness: Overly complex prompts can confuse AI. Short and direct prompts work best.
• Contextual Guidance: Including relevant background information helps the AI generate accurate responses.

Zero-Shot Prompting

Zero-shot prompting is a technique used in artificial intelligence, particularly with language models, where the model is asked to perform a task without being provided any specific examples or prior training data for that task within the prompt

No Prior Examples Needed

Maximizing AI Generalization

No Prior Examples Needed

Few Shot Prompting

Minimal Input for Complex Tasks

Few-shot prompting is a technique used in artificial intelligence, especially with language models, where the model is provided with a small number of examples or demonstrations of a task within the prompt.

Improved Flexibility

Adapting to Different Domains

Video created using D-iD and Canva.

Using Delimiters in Prompts

Using delimiters in prompts when engaging with generative AI significantly improves communication and task execution. Delimiters, such as quotation marks, brackets, or braces, help separate key instructions, ensuring the AI understands the task with precision. By using these clear markers, ambiguity is prevented, even in complex tasks. This structured approach enhances the accuracy of AI outputs, making it more likely to receive specific and accurate responses tailored to the given task.

QUESTIONS