Shadow AI Usage as an Emerging Insider Threat in Secondary Schools

Abstract

Because of pressures of work and the pace of work, education agencies are finding ways around networked, organizational IT controls and policies. The result is a fresh insider threat named Shadow AI. It has been reported that 66% of faculty members and 41% of students use artificial intelligence tools regularly without sanction. For instance, the secondary education sector is a critical security environment in which unauthorised generative AI systems have rapidly proliferated. This paper points out how the education policy does not lay enough focus on security vulnerabilities in regard to data vulnerabilities. The educational policy used to apply academic integrity bias prohibition as if they were stronger than the risks posed by educational cheating and educational cheating. Impulsive Behavior Students, Discretionary Workload Managing Educators, Negligent Insiders They skip and elude controls in the . For instance, an external organization obtained the release of confidential student records and personal information. The controls are unusable. The theft of guest information, use of third-party LLMs without permission, and invasion of privacy has become possible because of it. The review of literature identifies major areas for policy gaps. This policy is not very clear. It punishes too much. It misses basic AI education. The paper proposes governance frameworks to prevent the potential development of structural vulnerabilities at instances of the object that are ‘essentially safe’ and at instances that are ‘explicitly unsafe’, rather than at the proactive, reactive model.
There needs to be embedded AI literacy programmes, iterative policy review processes, and “privacy-by-design” controlled AI tools like licensed enterprise services and sandboxed educational tools for the successful integration of AI in secondary schools.