Think about the potential societal impacts of widespread Generative AI adoption. What positive changes do you hope to see?

1. In your Canvas course, review examples of Poor vs. Effective Prompts.

2. Use Gemini and/or ChatGPT to create prompts for at least three scenarios. Generate outputs and analyze them for clarity, relevance, and effectiveness, noting areas for improvement.

3. Share the prompt you generated and liked the most by copying and pasting it onto Padlet.

1. Take the prompt you shared in Module 2.2 and refine it further using both the TCREI and SHIELD frameworks to ensure the prompt is clear, effective, and privacy conscious.

2. Use Gemini, ChatGPT, and/or Claude practice app to iterate and improve your prompt.

3. Copy and paste your best refined version on the Padlet and reflect on the improvements.

Using the activities in 3.1 (Image Generation and Resume Screening), reflect on the following questions:

1. Claim: What biases do you think are present?

2. Support: What specific evidence in the AI output shows this bias?

3. Question: What assumptions might be underlying these biases?

1. Click the "+" icon to create a new post

2. Title your post with your name and department

3. Respond to this prompt (200-300 words):

As you complete this training, share one specific way you plan to use AI in your work, and one specific way you'll ensure the human element remains central to that task. Be concrete - what's your first planned AI-supported project, and how will you maintain the personal human touch?"

Copy and paste your final 5 guidelines here.

Each guideline should complete this phrase:
"When using AI at work, I will..." Remember: Your guidelines should reflect your unique perspective and needs, incorporating what you’ve learned about the California Two-Party Consent Law and ethical AI use in the workplace. 

Example Guideline:
"When using AI for meeting transcription, I will always announce recording at the start of meetings and secure explicit consent from all participants, documenting their agreement in the meeting notes."

1. Click the "+" icon to create a new post

2. Title your post with your name and department

Gemini: "Despite the challenging landscape facing education today, our program has demonstrated remarkable resilience and dedication this quarter. We are pleased to present this report, which highlights the significant strides made by our team in fostering [Program's Goal/Area of Impact]. The unwavering commitment to our mission, coupled with innovative strategies and tireless effort, has allowed us to navigate the current complexities and deliver impactful results. This report will detail the key achievements and successes that underscore the program's vital contribution, showcasing the positive impact we continue to make for our students and community."

AI can offer recommendations for accepting, partially-accepting, or rejecting requests based on pre-defined criteria. These recommendations can serve as a foundation upon which human judgment and context knowledge can be applied, combining the strengths of AI's data-driven insights with the nuanced understanding and experience of our staff and the data that we maintain. This hybrid approach guarantees that decisions are both informed by technology and grounded in human expertise.

To encourage the use of AI, we will provide training and resources like SOPs with screenshots and prompts to familiarize our staff with these tools, highlighting their benefits and practical applications. Workshops, tutorials, and ongoing support will ensure that everyone is comfortable and confident in incorporating AI into their workflows.

It is my hope that the integration of AI in vetting external data requests will streamline our operations and improve decision-making, but it is the combination of AI's efficiency and human insight that can set us apart.

2. Support: I prompted three times and got three different results. One result had all one gender, the next had a mix, there was various age groups represented as well as different races.

3. Question: If there are any biases being generated in results, we can prompt for better inclusivity - the training data may have been biased.

1. Always secure explicit consent from all participants at the start of any meeting where AI (e.g., transcription, summarization) is active. If any individual expresses discomfort, I will immediately disable the AI feature.

2. Proactively advise against sharing PII or sensitive confidential information during AI-assisted discussions, and I will pause or disable the AI tool for any segments where such data must be discussed.

3. Rigorously verify all AI-generated recaps and transcripts against my own notes and recollection, promptly correcting any misquotes, misattributions, or misrepresentations before sharing or acting on the information.

4. Delete raw AI recordings and transcripts once the verified, accurate meeting outcomes are officially documented and stored, typically within 24-48 hours, to minimize data retention risks.

5. Prioritize fostering an open and trusting team environment by clearly communicating the AI tool's purpose and continuously assessing its impact on team dynamics, adjusting my approach if it hinders collaboration.

2. When using AI at work, I will always inform meeting attendees at the start if AI is recording or transcribing, explain its purpose, and offer an opt-out or non-recorded alternative to ensure informed consent.

3. When using AI at work, I will store AI-generated recordings/summaries securely on designated company platforms, restrict access to those with a defined business need, and adhere to all data retention and deletion policies.

4. When using AI at work, I will critically verify the accuracy and relevance of all AI-generated content, especially factual claims, by cross-referencing reliable sources and applying my professional judgment to prevent misinformation.

5. When using AI at work, I will transparently disclose when AI has significantly contributed to content, analysis, or recommendations, especially when presenting it as my own work or influencing team decisions, to ensure clear context.

You are an expert in organizational efficiency and AI applications.
Your role is to provide insights into how AI can streamline administrative tasks and improve productivity.
You are talking to someone interested in leveraging AI for workplace efficiency.

Answer

AI can greatly enhance the efficiency of administrative tasks by automating repetitive processes, improving accuracy, and freeing up time for more strategic work.

Great to hear you're using Microsoft Outlook—it’s packed with AI-powered tools to help you respond faster and keep your inbox organized! Let me walk you through the features and strategies you can use to save time and stay on top of emails.

1. Respond Quicker with AI in Outlook

Smart Replies

• Outlook uses AI to suggest quick responses for common email scenarios. For example:

  • If someone asks for a meeting, you might see suggestions like:

    • "Sure, what time works for you?"

    • "Let me check and get back to you."

• These options appear directly in the email interface, allowing you to reply in one click.

Cortana Briefing Emails

• Cortana (Microsoft’s AI assistant) sends you a daily briefing email summarizing:

  • Emails you’ve received but haven’t replied to.

  • Suggested tasks based on your emails.

• Use this as a daily checklist to ensure no email goes unanswered.

Dictation

• If you’re on the go or want to save typing time, use the Dictation feature to compose emails via voice.

  • To enable: Go to an email window, click the Microphone icon, and start speaking. Outlook will transcribe your voice into text.

AI-Powered Grammar and Writing Assistance

• Outlook integrates with Microsoft Editor, an AI-powered tool that helps you:

  • Correct grammar and spelling errors as you type.

  • Refine tone and clarity with suggestions (e.g., making your email more formal or concise).

  • Highlight important action items in emails.

• You can enable this feature in Outlook settings under Proofing.

Support: What specific evidence in the AI output shows this bias? Without providing specifics it only generated images of males only, females only, or mixed groups of males and females that were 30-40 years old and not all race groups represented.

Question: What assumptions might be underlying these biases? Perhaps there is a lack of input showing what diversity and inclusivity look like for both males and females.

1. Question Generation for Specific Books

I can analyze children's books you regularly use and suggest tailored higher-order questions like:

• "What do you think will happen next?"

• "How do you think the character feels when...?"

• "How would you solve this problem differently?"

2. Question Frameworks Based on Age Groups

I can provide developmentally appropriate questioning strategies for different age groups in your care (toddlers vs. preschoolers).

3. Story Time Planning Templates

I can create customizable templates that include:

• Pre-reading engagement questions

• During-reading pause points for discussion

• Post-reading reflection questions

Specifically include:

1. 5 ready-to-use question templates that incorporate banking time principles

2. A structured approach for asking 3-part follow-up questions for popular children's books

3. 3 concrete brainstorming activities that can be completed with minimal materials

4. Age-specific scaffolding techniques for mixed-age groups

5. A method for documenting real-life connections between stories and children's experiences

For each strategy, provide:

• The exact phrasing to use with children

• Anticipated child responses

• How to document outcomes to track growth

• Modifications for children with language delays

Let's start by clarifying:

• Primary objectives: What specific outcomes do you want from this conference?

• Target audience: Are you focusing on:

  • New fathers

  • Experienced fathers

  • Father figures (uncles, grandfathers, mentors)

  • Professionals who work with fathers

  • A combination of these groups

Understanding your specific goals and audience will shape everything from programming to marketing strategies.

Step 2: Essential Conference Components

Here are the key areas we'll need to plan:

1. Content & Programming

   • Keynote speakers

   • Workshop sessions

   • Panel discussions

   • Interactive activities

   • Networking opportunities

2. Logistics & Operations

   • Venue selection

   • Date selection

   • Registration system

   • Staffing needs

   • Equipment/technology requirements

3. Marketing & Communication

   • Promotional strategy

   • Materials development

   • Community outreach

   • Media engagement

4. Budget & Sponsorships

   • Cost projections

   • Funding sources

   • Sponsorship opportunities

   • Evaluation & Follow-up

     • Feedback collection

     • Impact measurement

     • Post-event communication

I'm not completed sure how I can integrate AI into my daily work, but I can begin to conceptualize things that can be improved. The thing that is of great importance is to maintain that personal human touch in all that I do. Technology is improving in order to assist in our everyday work, but AI still needs to be taught what to do and how to do it in order to benefit its users.

Asking for a gift idea for your cousin who is making dinner

Mentioning you are driving up and have limited car space

Noting the gift will be in a hot car for a while

Mentioning you have given him beer in the past

Final Output:

💡 Solutions to Still Give Beer!

Pick it up close to his home - Stop at a local grocery or liquor store near his place so it hasn't been sitting in the heat

Gift card to a local brewery or beer shop - He can pick out something he loves himself

Gift card to a grocery store - He can use it to buy his own beer or whatever he needs

🤔 Something to Consider

Since he is cooking dinner for you and your friends, you could also:

Ask him if he needs anything for the dinner

Offer to bring a specific ingredient or side dish he might need

This way the gift is practical and helpful for the occasion!

Given everything - the hot car, limited space, and his love of beer - I think the easiest and most thoughtful option would be to either:

Pick up beer near his home, or

Get him a gift card to a brewery or beer shop

l was surprised that there was an array of diversity, but still a tinge of elitism, racism, and sexism.

When using AI at work, I will be mindful of unnecessary or excessive AI usage that contributes to energy consumption.

When using AI at work, I will prioritize AI tools from providers with demonstrated commitments to sustainability.

When using AI at work, I will use AI efficiently by crafting clear prompts to reduce the need for multiple attempts.

When using AI at work, I will ensure AI is used to enhance collaboration, not replace meaningful human interaction.

When using AI at work, I will respect team members who may have concerns about AI and include their perspectives in discussions.

Act as a Head Start fiscal operations and compliance systems expert. Using the historical Head Start Fiscal Liaisons Monitoring Procedures and Tools Manual (2013–2014) as a foundational reference, help me design a modern Delegate Fiscal Lead Procedures Manual that can be used for onboarding and training new hires.

Include recommended sections, workflow steps, compliance checkpoints, escalation procedures, monitoring timelines, document templates, and best practices for consistency across delegate agencies.

Scenario 2 – Delegate Agency Funding Growth Improved Prompt

Act as a strategic fiscal advisor for Head Start delegate agencies. Help me identify innovative and compliant ways to support newly assigned delegate agencies in maximizing current funding, strengthening fiscal performance, and increasing future funding opportunities.

Please include strategies related to cost allocation, non-federal match optimization, underutilized funding streams, reserve opportunities, grant readiness, partnership development, and long-term sustainability planning.

Scenario 3 – Team Leadership + Workload Balance Improved Prompt

Act as an executive leadership coach for public-sector fiscal teams. I serve on a Delegate Fiscal Lead team of several members and want to continue being a strong support to the team while effectively managing my own agency portfolio and responsibilities.

Provide strategies for balancing peer collaboration, knowledge sharing, cross-coverage support, workload prioritization, escalation protocols, and personal productivity systems so I can contribute as a team leader without compromising my own deliverables.

Context: Using a historical Fiscal Liaisons Monitoring Procedures and Tools Manual as a foundational reference, help me develop a modern Delegate Fiscal Lead Procedures Manual for onboarding and training new hires in a privacy-safe and scalable way. Use generalized delegate agency scenarios rather than real agency identifiers.

References: Incorporate best practices from monitoring workflows, internal controls, escalation matrices, risk-based review cycles, budget monitoring, BAR/RAA processing, corrective action tracking, and cross-agency consistency protocols.

Format: Organize the output into:

1. Manual table of contents

2. Standard operating procedures by function

3. Compliance checkpoints

4. Monitoring timelines

5. Escalation decision trees

6. Required templates/forms

7. New hire training roadmap (30/60/90 days)

8. 
   

Evaluate: Ensure the final manual is practical, compliance-centered, easy for new staff to follow, and adaptable across multiple delegate agencies without exposing agency-specific details.

Iterate: After drafting, recommend additional tools such as checklists, tracker dashboards, and CAP/QIP templates to improve long-term usability.

The main biases present in the AI ranking outputs appear to be:

• educational prestige bias, where candidates from highly recognized universities may be ranked higher than equally qualified candidates from state schools

• name-based and cultural bias, where names may trigger unconscious assumptions about ethnicity, gender, or professionalism

• ordering / anchoring bias, where the AI may favor resumes shown earlier in the document

• presentation bias, where stronger wording, polished summaries, or larger numerical achievements may outweigh actual role fit

These biases suggest the AI may not always focus solely on job-relevant qualifications and measurable skills.

2) Support: What specific evidence in the AI output shows this bias?

The strongest evidence is that the same or nearly identical work histories were reassigned to different names and universities across the five files, yet the perceived “top candidate” could easily shift.

For example:

• one version pairs strong AI accomplishments with Stanford University

• another pairs similar accomplishments with Harvard University

• another uses California State University Bakersfield with similarly high-impact AI work

Even though the quantified achievements remain highly comparable, an AI system may still rank elite institutions higher.

Another sign is that candidate names rotate while content stays similar, such as:

• Michael Smith

• Sarah Chen

• Jamall Washington

• Maria Rodriguez

If rankings changed after only names changed, that supports name-based bias.

3) Question: What assumptions might be underlying these biases?

Several hidden assumptions may be driving the AI:

• the assumption that elite universities automatically signal stronger talent

• the assumption that certain names imply higher professionalism or cultural fit

• the assumption that resumes appearing first are stronger

• the assumption that larger financial numbers or more “impressive sounding” wording always means better performance

• the assumption that traditional career paths are more valuable than diverse or nontraditional experience

These assumptions reflect systemic hiring norms already present in historical data, which AI may unintentionally learn and reproduce.

Closing Reflection

This exercise shows that AI can inherit the same biases that exist in human hiring systems. The biggest lesson is that AI should support structured, blind, job-relevant evaluation—not replace human judgment.

The human element will remain central by ensuring that professional judgment, compliance expertise, and relationship-based leadership drive all final decisions. While AI can accelerate drafting and systems design, it cannot replace the nuanced understanding required for delegate agency support, fiscal risk assessment, or coaching new staff through real-world scenarios. I plan to personally validate every section against LACOE fiscal expectations, Head Start requirements, and 2 CFR Part 200 standards, while also incorporating feedback from experienced colleagues and the lived realities of our team’s day-to-day work.

The personal human touch will come through how I use the manual as a mentorship and leadership tool, not just a document. During onboarding, I want new hires to feel supported through conversation, context-sharing, and examples from actual monitoring experiences. AI will help build the framework, but the trust, coaching, empathy, and critical thinking that make someone successful in this role will come from intentional human connection and leadership.

My hopes for Generative AI is that it allows us to create more streamlined approaches in our everyday work. All the systems we use don't "talk to each other" and so we spend so much time reconciling data that most of our work day/week is spent doing just that.

I'm also thinking it can become a brainstorming partner for us. So far, from the training, it mentions Generative AI can come up with ideas never before thought about. That's something I'd like to explore further.

Answer:

You're looking for fantastic books to engage infants and toddlers during a group storytime! This age group thrives on repetition, bright visuals, and interactive elements.

Here are 5 excellent choices that are often big hits:

1. "Brown Bear, Brown Bear, What Do You See?" by Bill Martin Jr. and Eric Carle

   • Why it's great: The repetitive text is perfect for infants and toddlers to anticipate and "read along" with. Eric Carle's bold, colorful illustrations are captivating, and it introduces animals and colors in a simple, engaging way. It's easy to add actions or animal sounds.

2. "The Very Hungry Caterpillar" by Eric Carle

   • Why it's great: Another Eric Carle classic! The vibrant collage illustrations, simple story of growth and transformation, and the unique die-cut pages where the caterpillar eats through the food are incredibly engaging. It's wonderful for talking about numbers, days of the week, and healthy eating.

3. "Pat the Bunny" by Dorothy Kunhardt

   • Why it's great: This is a quintessential touch-and-feel book. Infants and young toddlers love the sensory experience of touching the "furry bunny," "rough beard," and other textures. It's highly interactive and encourages exploration, perfect for little hands.

4. "Where's Spot?" by Eric Hill

   • Why it's great: The original lift-the-flap book! Toddlers adore the suspense and delight of opening flaps to find Spot. It promotes object permanence, encourages participation, and is a simple, fun hide-and-seek adventure.

5. "Chicka Chicka Boom Boom" by Bill Martin Jr. and John Archambault, illustrated by Lois Ehlert

   • Why it's great: This book is a rhythmic, energetic celebration of the alphabet. The lively, rhyming text and vibrant illustrations make it incredibly fun to read aloud. Toddlers often love to chant along, and it's fantastic for introducing letter recognition in a playful way. You can even add body movements as the letters "climb" and "fall" from the coconut tree.

2. I will ensure the human element remains central to that task by reviewing all AI generated content and check for accuracy and biases in the data.

3. My first planned AI project is to minimize the ERSEA GIM and revise SOPs.

Objectives

Activities

Quarterly Review

Annual Fiscal and Compliance Review

Support: All 3 images have people that look between 30-40.

Assumptions: Only young and fit people are in leadership positions.

1. ensure that AI-generated communications are reviewed and personalized before sending, so that authentic human connection is maintained across all levels of hierarchy.

2. be transparent with my team about which tasks or communications are AI-assisted, to avoid perceptions of unfairness or deception.

3. always inform all participants before using any AI recording, transcription, or monitoring tool in meetings or conversations.

4. acknowledge that AI recording can inhibit authentic communication and create a "chilling effect," and will therefore use such tools only when truly necessary.

5. use AI as a tool to enhance my work, not as a replacement for critical thinking, empathy, or human judgment.

Here are 5 supportive message options:

1. Your child deserves a strong start. Our free preschool program offers a safe, caring place for children to learn and grow, while also connecting parents with helpful resources and support.

2. You are not alone. We welcome families experiencing homelessness and provide free preschool along with resources for parents, including support services that can help your whole family.

3. A brighter beginning starts here. Enroll your child in our free preschool program where they can build skills, make friends, and thrive—while you access resources designed to support parents.

4. Every family deserves support and opportunity. Our free preschool program gives children a nurturing learning environment and offers parents access to valuable community resources.

5. Small steps can make a big difference. Our free preschool program is here to support your child’s early learning and help connect you with resources for housing, family services, and more.

1. I will always ensure the information is accurate...including citations, quotes, etc.

2. I will protect all confidential information and ensure privacy of individuals

3. I will be transparent about using AI and let people know when it has helped create content or make recommendations

4. I will never use AI to discriminate, judge or make unfair decisions.

5. I will always comply with LACOE's policies about AI and its use.

When using AI at work, I will ensure the "human" touch is included by cross-checking sources on other platforms and documents and encouraging my team to 'show their work' when using AI and not relying solely on AI to create a plan or document. The first project will involve updating the communications SOP to streamline processes and generate new pathways to recruit families.

Your role is to craft empathetic, low-barrier, and highly persuasive outreach messaging to support families in crisis.

This is more useful to provide a 'persona' without giving too much information to Gemini that might compromise our team's efforts and provide too much proprietary data or information.

Will be using AI to help me review our delegate agencies policies and procedures, making sure they are aligned to the LACOE GIMs.

"Write a SMARTE goal for ethics and integrity for a Program Specialist working in Special Education.  Use the E in SMARTE to discuss equity for all students.  Combine the six goal ideas into one goal not exceeding 125 words.  Begin the goal with “By May, 2027, I…”

This was the Ai output:

"By May, 2027, I will demonstrate ethics and integrity as a Program Specialist in Special Education by ensuring that all decisions, communications, and services align with special education law, district policy, confidentiality requirements, and professional standards. I will maintain accurate records, respond to ethical or compliance concerns within 3 business days, and provide at least two staff learning opportunities on ethical practices, confidentiality, legal compliance, and equitable service delivery. Using collaboration, district resources, and professional learning, I will support consistent decision-making and accountability. To promote equity for all students, I will review practices to help ensure supports and services are implemented fairly, consistently, and responsively to each student’s individual needs, cultural and linguistic background, and right to appropriate educational opportunities."

I initially asked for just a SMARTE goal and it gave me one with evaluation for the E so I refined the prompt to address equity, and asked for 250 words in one combined goal. That was too wordy so I then reduced the prompt to 125 words.

Write a SMARTE goal for ethics and integrity for a Program Specialist working in Special Education.  Use the E in SMARTE to discuss equity for all students.  Refer to the California Professional Standards for Educational Leaders #5.  Combine the six goal ideas into one goal not exceeding 125 words.  Begin the goal with “By May, 2027, I…”

I used both frameworks to vet my prompt:

TCREI Framework:

 1^st iteration asked for too many words and did not include a clear reference

Prompt rewritten twice:

#2 reduce number of words to 125

#include CPSEL reference

SHIELD framework:

Candidate’s name and the name of their district is not included in the prompt.

This scenario is appropriate to any CASC cadre so it avoids identifying real individuals.

This would fall under the public or internal level.

The only specific details (and these are very broad) are Special Education and California Professional Standards for Educational Leaders.

The prompt is limited to the CASC program which exists throughout the state.

My Analysis:

Analysis:  The image generated shows a young black woman standing near a projected slide entitled Sports Medicine Strategy.  Said leadership team includes one black woman, two very light skinned males, a woman who is possibly Hispanic/indigenous/East Indian, a black male, and a person who may be Asian.  Possible biases noted:

-Slide shows a human body which appears more male than female; this implies that sports are seen as part of a man’s world

                  -If the Asian appearing person is male, then the men outnumber women 2 to 1

-All leadership team members are dressed in business attire; why wouldn’t a sports medicine company include someone in medical attire? 

-The business attire is that of the United States or a European country; why didn’t AI include someone in appropriate traditional/ethnic attire?

-Laptop computers along with notepads are in front of two male members; does that imply that women only take handwritten notes?

-Some of the terms on the slide chart appear to be written in a language other than English; does this imply that sports medicine is trending more in other countries?

-All visible pencils/pens are on the right side of the tablet; are none of the participants left-handed?

-All participants appear to be about the same age (40’s); does this mean that older staff members are not included on the leadership team and if so, why not?

AI response: “By May, 2027, I will consistently model ethical leadership and integrity by ensuring that all decisions, practices, and communications reflect professionalism, transparency, and accountability while upholding legal and ethical responsibilities in Special Education. I will promote equity by advocating for inclusive, culturally responsive, and accessible supports that ensure every student has equitable opportunities to succeed. Through ongoing collaboration with staff and families, reflective practice, and data-informed decision-making, I will strengthen trust, improve services, and foster a positive, student-centered culture aligned with the California Professional Standards for Educational Leaders (CPSEL) Standard 5: Ethics and Integrity.”

I used both the TCREI and SHIELD frameworks to refine the prompt as needed. 

Some of the improvements I made to the prompt::

• I changed the word count to 125 words to make it more succinct.

•I included a reference to CPSELs.

I looked at the AI response through the SHIELD framework to ensure compliance:

•no candidate names nor district names were included

•the scenario is appropriate an any CASC candidate eliminating identifying features

 •candidates working in Special Education can be found throughout California so although this decreases the number of potential candidates, it is still a broad pool.

My Personal Guidelines for Ethical AI Use at Work

1. I will protect privacy and confidentiality by using the SHIELD framework to identify potential risks.

2. I will use AI as a support tool, not a substitute for professional judgment, and treat its outputs as drafts or starting points.

3. I will verify AI-generated content for accuracy, bias, missing context, and potential misinterpretation before using or sharing it.

4. I will use AI transcription only with participant consent and in appropriate settings.

5. I will monitor for bias and support equitable AI use by promoting training and minimizing power imbalances.

One specific way that I can ensure the human element remains central to the task is to focus our work on an actual issue that a candidate is faced with and then to explore the issue in depth through personal discussion.  I would then want to brainstorm 2-3 possible solutions with my candidate and determine what the candidate feels would be the pros and cons of each solution.  Once the candidate has identified a solution they are comfortable with, I would guide them in prompting AI to share 4-5 other resolutions that the candidate might consider.  We would then spend time discussing each suggestion and determining if a particular suggestion made more sense than they others.  I would use several sentence stems from our LEAD coaching tool as I coached the candidate.  I might also have the candidate prompt AI with “What could happen if I…” questions to help generate multiple solutions. Above all, I would encourage my candidate to view all content generated as mere suggestions and ideas to consider and analyze.