AI Use Policy
Last updated: 2026-05-29 · Version 1.0
Grade Harbor LLC (“Grade Harbor,” “we,” “us,” or “our”) provides a teacher-facing educational technology service that helps teachers review, grade, and provide feedback on student writing.
This AI Use Policy explains how Grade Harbor uses artificial intelligence in its website, application, AI-assisted grading tools, feedback tools, and related services collectively referred to as the “Service.”
1. How Grade Harbor Uses AI
Grade Harbor uses artificial intelligence to help teachers review student writing more efficiently and consistently.
The Service may use AI to:
- Analyze student writing against teacher-selected or school-approved rubrics;
- Generate per-criterion scoring suggestions;
- Suggest score ranges;
- Draft written feedback for teacher review;
- Identify writing strengths and areas for improvement;
- Surface instructional writing skill insights;
- Help teachers compare student work to rubric expectations;
- Support grading workflow efficiency; and
- Assist with drafting comments, explanations, or instructional next steps.
AI-generated output is intended to support teacher judgment. It is not intended to replace professional review by a teacher or other qualified educational professional.
2. How Grade Harbor Does Not Use AI
Grade Harbor does not use AI to make final grading decisions.
Grade Harbor does not:
- Automatically assign final grades without teacher review;
- Automatically send AI-generated feedback directly to students;
- Replace teacher judgment;
- Make disciplinary decisions;
- Make placement, promotion, retention, or eligibility decisions;
- Use Student Data for targeted advertising;
- Use Student Data to build non-educational profiles;
- Sell or rent Student Data;
- Use Student Data for unrelated commercial purposes; or
- Use Student Data to train generalized AI models without explicit written authorization from the applicable school, district, or authorized educational institution.
AI-generated suggestions must be reviewed and approved by a teacher before they are treated as final.
3. Teacher Review and Human Oversight
Human review is central to Grade Harbor's design.
Teachers are responsible for:
- Reviewing AI-generated grading suggestions;
- Reviewing AI-generated feedback;
- Correcting inaccurate or incomplete output;
- Adjusting scores where appropriate;
- Confirming that feedback is age-appropriate and instructionally useful;
- Ensuring that final grades align with classroom expectations, school policy, and professional judgment; and
- Approving all final grades, scores, comments, and feedback before they are shared or relied upon.
Grade Harbor is an assistive tool. The teacher remains the decision-maker.
4. AI-Generated Grades and Feedback
AI-generated grades, rubric scores, comments, and feedback are draft suggestions.
They may be helpful, but they may also be:
- Inaccurate;
- Incomplete;
- Inconsistent;
- Too harsh or too generous;
- Misaligned with a teacher's intent;
- Misaligned with a school's grading policy;
- Affected by unclear rubric language;
- Affected by missing assignment context; or
- Otherwise inappropriate for a particular student or classroom.
Teachers should review AI-generated output carefully before using it.
If an AI-generated suggestion is incorrect, incomplete, inappropriate, or inconsistent with teacher judgment, the teacher should edit, override, or reject it.
5. Student Data and AI Processing
To provide AI-assisted grading and feedback, Grade Harbor may process student-related information submitted by teachers, schools, districts, or other authorized educational users.
This may include:
- Student writing or essay text;
- Assignment prompts;
- Rubrics;
- Teacher instructions;
- Draft scores;
- Draft feedback;
- Class, course, or assignment context; and
- Other educational materials submitted through the Service.
Grade Harbor does not require student names, student email addresses, student phone numbers, student photos, or student account credentials to provide AI-assisted grading and feedback.
Because student writing and uploaded materials may contain free-form text, those materials may accidentally include personal information that Grade Harbor did not request. Teachers and schools should avoid uploading unnecessary personal information and should remove student names or direct identifiers where practical.
6. AI Model Training
Grade Harbor does not use Student Data to train generalized AI models without explicit written authorization from the applicable school, district, or authorized educational institution.
Grade Harbor does not knowingly opt in to third-party AI provider model-training data sharing for Student Data unless the applicable school, district, or authorized educational institution has provided explicit written authorization.
Grade Harbor may use de-identified or aggregated information to improve the reliability, security, usability, and performance of the Service, as described in our Privacy Policy and any applicable written agreement. De-identified or aggregated information is not intended to reasonably identify an individual student, teacher, school account, or household.
7. AI Service Providers
Grade Harbor may use third-party AI service providers to help deliver AI-assisted grading, rubric analysis, feedback drafting, and related functionality.
Grade Harbor does not publish a full AI provider or subprocessor list on this page. Where required by a school, district, applicable law, or signed agreement, Grade Harbor can provide additional service provider information during a privacy, procurement, security, or Data Processing Agreement review.
Grade Harbor does not authorize AI service providers to use Student Data for advertising, resale, or unrelated commercial purposes.
8. Bias, Fairness, and Rubric Quality
AI-assisted grading depends heavily on the quality of the rubric, assignment prompt, teacher instructions, and student writing submitted to the Service.
AI systems may produce output that reflects limitations, assumptions, or patterns in the underlying model, rubric, prompt, or training data. AI output may also be affected by unclear grading criteria, ambiguous writing prompts, or incomplete classroom context.
Teachers should review AI-generated suggestions for fairness, accuracy, and appropriateness.
Teachers and schools should consider:
- Whether the rubric is clear and age-appropriate;
- Whether grading criteria are aligned to the assignment;
- Whether the AI feedback is instructionally useful;
- Whether score patterns appear unusual or inconsistent;
- Whether students with different writing styles are being evaluated fairly; and
- Whether final feedback reflects teacher judgment and classroom expectations.
Grade Harbor does not guarantee that AI-generated output will be free from bias or error.
9. Appropriate Use of AI Output
Teachers should use Grade Harbor's AI output as a support tool for educational purposes.
Teachers should not use AI-generated output as the sole basis for:
- Final grades without review;
- Student discipline;
- Special education determinations;
- Promotion or retention decisions;
- High-stakes placement decisions;
- Eligibility decisions;
- Legal, medical, or psychological conclusions; or
- Any decision that requires independent professional judgment beyond classroom writing assessment.
Grade Harbor is designed to assist with writing feedback and grading workflows. It is not designed to make high-stakes decisions about students.
10. No Direct Student Communication
Grade Harbor is designed as a teacher-facing tool.
Students do not create Grade Harbor accounts, and Grade Harbor does not knowingly collect information directly from students.
AI-generated feedback is provided to the teacher for review. Grade Harbor does not automatically communicate AI-generated feedback directly to students.
11. Accuracy and Limitations
Grade Harbor works to make AI-assisted grading and feedback useful, consistent, and aligned with teacher-selected rubrics. However, AI-generated content may not always be accurate, complete, appropriate, or consistent.
Grade Harbor does not guarantee that:
- AI-generated scores will be correct;
- AI-generated feedback will be complete;
- AI-generated feedback will be appropriate for every student;
- AI-generated output will match every teacher's grading style;
- AI-generated output will satisfy every school, district, or state grading requirement;
- AI-generated output will identify every writing issue; or
- AI-generated output will be error-free.
Teachers are responsible for reviewing, editing, approving, or rejecting AI-generated output before relying on it.
12. Changes to AI Use
Grade Harbor may update its AI features, AI workflows, model configurations, or AI service providers from time to time.
If we make material changes to how Grade Harbor uses AI with Student Data, we will notify account holders by email or through the Service at least 30 days before the changes take effect, unless a shorter period is required for legal, security, or operational reasons.
Where required by a signed school or district agreement, Grade Harbor will provide notice or obtain approval in accordance with that agreement.
13. Relationship to Other Grade Harbor Policies
This AI Use Policy should be read together with Grade Harbor's:
- Privacy Policy;
- Terms of Service;
- FERPA & COPPA Notice;
- Data Processing Agreement request page;
- Any signed school, district, or data processing agreement; and
- Any applicable order form or procurement document.
If a signed agreement with a school or district provides additional protections or different requirements, that signed agreement will control to the extent stated in the agreement or required by applicable law.
14. Contact
For questions about Grade Harbor's use of AI, student data, privacy, or school review processes, contact:
Grade Harbor LLC
Email: [email protected]