IB ACADEMIC INTEGRITY AND AI: A CLASSROOM CHECKLIST FOR TEACHERS

Define AI use at the task level

A department-wide policy establishes principles, but students also need a short statement on each significant task. It should say whether AI tools may be used, for what purpose, what must be cited or described, what data must not be entered and which parts must be completed without assistance. Examples reduce ambiguity.

The IB’s public guidance recognizes possible learning benefits while emphasizing critical review, bias awareness and appropriate credit for AI-generated work. Teachers can therefore design uses that deepen inquiry—such as critiquing an output—without treating generated text as the final demonstration of student achievement.

Collect evidence of thinking instead of hunting for a detector score

Authentic process evidence is valuable even when no AI concern exists. Short oral explanations, planning notes, source trails, drafts, data decisions and reflections show how understanding develops. They also let a teacher adapt instruction before the final submission.

A detector output can be wrong and does not explain the learning process. If work creates a concern, use the school’s procedure, examine multiple sources of evidence and give the student an opportunity to explain the work. The response should be proportional, documented and consistent.

Run a privacy and equity check before adopting a tool

A tool that is pedagogically interesting may still be unsuitable if it collects student data, requires an inappropriate account, lacks accessibility or advantages students who can pay. Schools should review age terms, data handling, retention, model training, security and approved procurement before requiring use.

Provide an equivalent route for students who cannot or should not use the tool. Evaluate whether the activity improves the intended learning outcome, not only whether students enjoy it. Retire uses that add surveillance, confusion or workload without better evidence of learning.

Classroom principle. Use AI when it makes student thinking more active and visible. Avoid uses that outsource the thinking the task is supposed to assess.

• Do not enter personal, confidential or unpublished student work into an unapproved service.
• Check age, consent, retention and account requirements before classroom use.
• Provide a non-AI route that assesses the same learning objective.
• Teach students to verify claims and trace primary sources.
• Document high-stakes decisions and keep a responsible human decision-maker.
• Review the practice regularly as tools, policy and evidence change.

Continue reading