## Ethical Considerations and Safe Practices

### **Responsible AI Use in Educational Settings**

Implementing AI ethically requires ongoing attention to privacy, bias, academic integrity, and student development. This isn’t about restricting technology—it’s about using it thoughtfully.

#### **Student Privacy and Data Protection**

**Understanding Data Collection** Most AI tools collect and store user interactions. This can include:

- Text inputs (prompts and questions)
- Generated outputs and responses
- Usage patterns and frequency
- Account information and preferences

**FERPA Compliance Considerations**

- Never input personally identifiable student information into AI tools
- Avoid uploading student work directly to AI platforms
- Use generic examples instead of specific student scenarios
- Check if your district has approved specific AI tools for educational use

**[Safe Practices for Protecting Student Information](/content/post/7-ai-guidelines-your-students-actually-want-to-follow/index.html)**

- Create anonymized scenarios for AI analysis
- Use hypothetical student examples rather than real ones
- Focus AI assistance on curriculum content rather than individual student data
- Teach students about digital privacy and data protection

**Example of Safe vs. Unsafe AI Use:**

**Safe:** “Create a math word problem about a student who saves money for a bicycle”  
**Unsafe:** “Help me write an email to Maria’s parents about her behavioral issues in class”  
**Safe:** “Generate discussion questions about this historical event: [paste event description]”  
**Unsafe:** “Analyze this student’s essay and tell me what grade to give: [paste student work]”

#### **Addressing Bias and Ensuring Fairness**

**Understanding AI Bias** AI systems reflect the biases present in their training data, which can include:

- Gender stereotypes (assuming certain careers are “for boys” or “for girls”)
- Cultural assumptions (defaulting to certain cultural perspectives)
- Socioeconomic bias (assuming access to resources not all students have)
- Language bias (performing better with standard English than dialects)

**Creating Inclusive AI Experiences**

**Diverse Prompting Strategies:**

- Request examples that include various cultural backgrounds
- Ask for scenarios that represent different family structures
- Specify the need for inclusive language and representation
- Vary names and pronouns in examples to reflect classroom diversity

**Critical Evaluation Practices:**

- Review AI outputs for missing perspectives
- Check for stereotypical assumptions
- Ensure examples are accessible to students with different abilities
- Include student voices in evaluating AI-generated content

**Teaching Students About Bias:**

- Discuss how AI learns from human-created data
- Practice identifying bias in AI responses
- Compare AI outputs with diverse sources
- Encourage questions about whose perspectives are represented

#### **Academic Integrity in the AI Era**

**Redefining Original Work** Traditional definitions of plagiarism need updating for the AI age. Consider these principles:

- Students should understand and be able to explain their work
- AI assistance should enhance learning, not replace it
- Transparency about AI use should be expected and documented
- The focus should be on learning objectives rather than just end products

**Developing AI Use Policies**

**Appropriate AI Use Examples:**

- Brainstorming ideas for projects or essays
- Getting explanations of difficult concepts
- Grammar and style suggestions for writing
- Creating outlines or organizing thoughts
- Generating practice problems or study materials

**Inappropriate AI Use Examples:**

- Having AI complete entire assignments without student input
- Copying AI responses without attribution or understanding
- Using AI for assessments meant to measure individual knowledge
- Submitting AI-generated work as original student creation

**Documentation Requirements:** Students should be able to explain:

- When and how they used AI assistance
- What AI tools they used and for what purposes
- How they verified or built upon AI suggestions
- What they learned through the AI interaction

#### **Maintaining Student Agency and Learning**

[**Keeping Students at the Center** AI should amplify student thinking, not replace it](/content/post/ai-and-critical-thinking-finding-the-right-balance/index.html). Strategies include:

- Using AI as a starting point for student exploration
- Using AI as a Socratic coach, not as a shortcut that does the work
- Requiring students to critically evaluate AI suggestions
- Emphasizing the importance of student voice and perspective
- Teaching students when NOT to use AI

**Teaching Critical Evaluation Skills**

**Questions Students Should Ask About AI Outputs:**

- Does this information seem accurate based on what I know?
- What perspectives might be missing from this response?
- How can I verify this information through other sources?
- Does this response actually answer my question?
- What assumptions is the AI making that I should examine?

**Balancing AI Assistance with Skill Development**

- Provide regular opportunities for independent work
- Teach students to use AI as a Socratic coach that asks questions, challenges assumptions, and helps them think deeper
- Teach multiple problem-solving strategies beyond AI assistance
- Emphasize process over product in learning activities
- Help students develop metacognitive awareness of their learning

#### **[Grade-Level Considerations for Ethical Implementation](/content/post/ai-fact-checking-101-teaching-students-to-verify-not-just-trust/index.html)**

**Elementary (K-5): Foundation Building**

- Focus on teacher-mediated AI use
- Introduce concepts of “computer helpers” and their limitations
- Teach basic digital citizenship and privacy awareness
- Emphasize human creativity and critical thinking

**Sample Elementary Lessons:**

- “Why we need to check what computer helpers tell us”
- “Keeping our personal information private”
- “When to ask a human for help instead of a computer”

**Middle School (6-8): Critical Thinking Development**

- Begin supervised direct student use of AI tools
- Introduce concepts of bias and fairness in technology
- Develop guidelines for appropriate vs. inappropriate use
- Start conversations about AI’s impact on society

**Sample Middle School Discussions:**

- “How might AI be biased and how can we recognize it?”
- “What makes something ‘original work’ when AI tools exist?”
- “How do we balance AI assistance with developing our own skills?”

**High School (9-12): Preparation for Independence**

- Encourage independent, responsible AI use with clear boundaries
- Explore complex ethical questions about AI in society
- Develop sophisticated critical thinking about AI capabilities and limitations
- Prepare students for AI use in college and careers

**Sample High School Projects:**

- Research paper on AI ethics with analysis of multiple perspectives
- Debate about AI’s role in future careers and society
- Creation of AI use guidelines for incoming students
- Analysis of bias in AI systems with proposed solutions

#### **Developing School-Wide Policies**

**[Key Policy Components](/content/post/how-schools-addressed-privacy-concerns-when-integrating-ai-tools/index.html):**

- Clear definitions of appropriate and inappropriate AI use
- Procedures for documenting AI assistance in student work
- Guidelines for protecting student privacy and data
- Professional development requirements for staff
- Regular review and update procedures as technology evolves

**Implementation Strategies:**

- Start with pilot programs in interested classrooms
- Provide ongoing professional development for all staff
- Include parents and students in policy development conversations
- Create feedback mechanisms for continuous improvement
