Responsible AI Policy

CEAIH AI tools are intended to support instruction, tutoring, planning, feedback, accessibility, content creation, administration, and analysis. They are not a replacement for qualified educators or other professionals.

• Teachers approve individualized remarks and consequential instructional decisions.
• Authorized personnel review grades, discipline, placement, accommodations, and official reports.
• Users verify factual claims and appropriateness before publishing or relying on AI output.
• Organizations remain responsible for decisions made using CEAIH.

CEAIH will seek to identify AI-assisted features and provide reasonable information about their purpose. Users should disclose AI assistance when required by school policy, academic rules, law, or professional standards.

AI can make mistakes, invent facts, misunderstand context, reproduce bias, or generate inappropriate material. Output should be treated as a draft or recommendation unless independently verified.

CEAIH aims to support Caribbean languages, curricula, cultures, accessibility needs, and diverse educational contexts. CEAIH will evaluate and improve AI features to reduce unfair bias and exclusion.

Users should provide only information necessary for the task. Sensitive student, health, legal, financial, biometric, authentication, or security information should not be entered into general AI prompts unless an approved workflow specifically requires and protects it.

• Impersonation, fraud, harassment, exploitation, or deception.
• Generating sexual content involving minors or facilitating child exploitation.
• Creating malware, evading security, or enabling unauthorized access.
• Discriminatory profiling or automated exclusion based on protected characteristics.
• Making final high-impact decisions without authorized human review.
• Academic cheating or misrepresenting AI-generated work as solely one’s own where prohibited.

CEAIH may apply automated and human moderation, age-appropriate restrictions, usage limits, logging, reporting, and account action to protect users and the platform.

Users must respect copyrights, trademarks, privacy, publicity, and other rights when requesting or using generated content. AI output may resemble existing content and may not be unique.

CEAIH will review feedback, incidents, educational outcomes, provider changes, and evolving standards to improve AI safety and effectiveness.