AI CYBER SAFETY TOOLKIT
Responsible AI tools for safer, more resilient schools
The CYBERSAFE AI-supported Cyber Safety Toolkit is the core technological result of the project.
It is designed to help schools move from mainly reactive responses to digital incidents towards a more proactive, prevention-oriented approach. The toolkit combines interactive education, realistic digital-risk scenarios and anonymised school-level insights to support teachers and students in understanding and addressing cyberbullying, grooming, sexting and other emerging online risks.
WHAT IS THE CYBERSAFE TOOLKIT?
The toolkit will consist of three interconnected components:
1 — EDUCATIONAL MOBILE APPLICATION
A digital learning environment for students and teachers.
The application will provide:
age-adapted educational content;
interactive learning modules;
scenario-based activities;
structured guidance on safe and responsible digital behaviour;
secure access for participating users.
It is intended for use across primary, secondary and upper-secondary education, allowing activities to be adapted to different ages and educational contexts.
Expected result:
A fully functional educational application ready for pilot testing in participating schools.
2 — AI-DRIVEN VIRTUAL CASE STUDY GENERATOR
Learning through realistic situations
The Virtual Case Study Generator will create realistic and age-appropriate digital-risk scenarios for use in classroom activities.
Scenarios may address situations such as:
Cyberbullying
Online grooming
Sexting and non-consensual sharing
Online manipulation
Privacy and data misuse
Unsafe online behaviour
Teachers will be able to use these cases to stimulate discussion, critical thinking and guided reflection.
The purpose is not simply to tell students what is right or wrong, but to enable them to analyse situations, understand consequences and discuss possible responses in a safe educational environment.
EDUCATIONAL APPROACH
Scenario → Discussion → Reflection → Decision-making → Prevention
3 — CYBER SAFETY ANALYSIS ENGINE
From anonymous information to better prevention
The Cyber Safety Analysis Engine will generate anonymised insights and structured reports on recurring digital-risk patterns.
The system is intended to help teachers identify common themes and adapt prevention strategies accordingly.
For example, aggregated information may help a school understand whether participating groups need additional educational activities relating to:
privacy;
cyberbullying;
online manipulation;
inappropriate content sharing;
responsible communication.
The analysis is intended to support teachers and schools at class or institutional level, not to evaluate individual students.
HOW THE TOOLKIT WORKS
LEARN
Students engage with interactive digital-safety content.
→
EXPERIENCE
AI-generated case studies present realistic situations.
→
REFLECT
Teachers guide discussion and structured classroom activities.
→
UNDERSTAND
Anonymised information helps identify recurring risk themes.
→
PREVENT
Schools adapt educational and safeguarding strategies.
The overall logic of the toolkit is to support awareness, prevention and early identification of digital-risk themes, while strengthening schools’ ability to integrate digital safety into everyday educational practice.
RESPONSIBLE AI BY DESIGN
AI FOR EDUCATION — NOT SURVEILLANCE
CYBERSAFE is based on a Privacy by Design approach.
The AI-supported system is intended to strengthen educational prevention and school safeguarding while respecting privacy, ethical AI principles and GDPR requirements.
NO INDIVIDUAL PROFILING
The toolkit is not designed to create individual behavioural profiles of students.
NO INDIVIDUAL RISK SCORES
Students will not receive automated risk classifications or behavioural scoring.
NO AUTOMATED DISCIPLINARY DECISIONS
AI outputs will not be used to make automated disciplinary or punitive decisions.
NO DIRECTLY IDENTIFIABLE STUDENT DATA
The project foresees the use of anonymised identifiers rather than directly identifiable personal student data.
PRIVACY, SAFEGUARDING & ETHICS
Technical development will prioritise:
Anonymised student identifiers
GDPR-compliant data handling
EU-based hosting
Data sovereignty
Bias monitoring
Pedagogical safety
Responsible AI use
The Application also assigns an Ethics Officer from DDE Viotias to monitor AI outputs for bias and pedagogical safety and to support alignment with European ethical principles for AI in education.