STRIDE Analysis and Protective Mechanisms for Educational Chatbots

The article presents an analysis of contemporary trends in the use of educational chatbots, highlighting various aspects of demand for these technologies in the education sector. It is noted that despite significant advantages associated with the use of chatbots, there are risks concerning their security. The application of the STRIDE methodology allowed for the identification and analysis of potential security threats, such as spoofing, falsification, attacks on confidentiality, and data availability. The authors have identified key elements of the structure of educational chatbots that require protection. Proposed security measures correspond to identified threats and include authentication, end-to-end encryption, the use of API with authentication and keys, as well as self-destructing messages. The effectiveness of the proposed solutions was evaluated on the basis of simulation tests in simulated educational environments, which made it possible to determine the level of neutralization of threats, the impact on system performance and adaptability to new types of attacks. Additionally, the article addresses the issue of protection against denial of service attacks and privilege escalation. Possibilities for further research are discussed, pointing out the necessity of developing specialized threat models for educational chatbots, enhancing protection tools, and considering new technologies such as artificial intelligence and machine learning. The presented study attests to the effectiveness of the STRIDE methodology in identifying threats and justifying protective measures for educational chatbots. The proposed protection mechanisms make a significant contribution to the development of educational chatbot security; however, the work also indicates the need for further research to explore new methods and technologies to ensure security and adapt to growing threats.