Identifying and Managing AI-Related Risks
This intervention framework references recognised UK education guidance — including Department for Education (DfE) generative AI guidance for schools, Information Commissioner's Office (ICO) expectations on pupil data, Keeping Children Safe in Education (KCSIE), Joint Council for Qualifications (JCQ) and Ofqual assessment integrity guidance, and Ofsted expectations for safeguarding and curriculum quality. It supports school self-assessment; it is not legal advice and does not imply official endorsement or compliance certification.
Why This Matters
Every school manages risk.
AI introduces new opportunities, but also new forms of exposure.
An AI Risk Register helps schools:
- Identify risks
- Assign ownership
- Track controls
- Monitor improvement
This register should be reviewed regularly alongside safeguarding, data protection and strategic risk reviews.
Risk Category 1: Data Protection
Example Risk
Staff enter identifiable pupil information into public AI tools.
Potential Impact
- Data breach
- Loss of trust
- Regulatory concerns
Example Controls
- Staff training
- Approved tools list
- Privacy guidance
Suggested Owner
DPO / AI Lead
Risk Category 2: Safeguarding
Example Risk
AI-generated content contributes to bullying, impersonation or deepfake incidents.
Potential Impact
- Pupil harm
- Safeguarding concerns
Example Controls
- Online safety education
- Staff awareness
- Incident reporting procedures
Suggested Owner
DSL
Risk Category 3: Assessment Integrity
Example Risk
Students submit AI-generated work without disclosure.
Potential Impact
- Reduced learning
- Academic dishonesty
Example Controls
- Assessment review
- Student guidance
- Teacher verification processes
Suggested Owner
Assessment Lead
Risk Category 4: Staff Over-Reliance
Example Risk
Staff accept AI-generated outputs without sufficient review.
Potential Impact
- Inaccurate teaching materials
- Reduced professional judgement
Example Controls
- Verification framework
- Human oversight expectations
- Professional development
Suggested Owner
AI Lead
Risk Category 5: Governance
Example Risk
AI use expands without oversight or policy review.
Potential Impact
- Inconsistent practice
- Increased exposure
Example Controls
- AI policy
- Annual review
- Governor reporting
Suggested Owner
Headteacher / Trust Lead
Risk Matrix
| Risk | Likelihood | Impact | Owner | Review Date |
|---|---|---|---|---|
| Data Protection | Medium | High | DPO | Annual |
| Deepfakes | Medium | Medium | DSL | Annual |
| Assessment Integrity | High | High | Assessment Lead | Annual |
Review Questions
- Has the risk changed?
- Are controls effective?
- Has training been completed?
- Are new risks emerging?
Improve Your Benchmark Score
Supports:
- Governance
- Privacy
- Safeguarding
- Assessment Integrity
Effective risk management is continuous, not one-off.
Retake the leader benchmark after updating your risk register
Wider Support & Development
For more information, guidance and support — including how AI Awareness Day can help your school — contact the AI Awareness team.