Solution Brief Intel Security
Security in Education AI and Confidential Computing help make secure remote exams a reality Accessibility, Data Security, and Fraud Detection for Distance Learning As an e-university where 80% of classes are delivered over the Internet, the Swiss Distance University of Applied Sciences (FFHS) takes online privacy and security very seriously. Even with more than twenty years of experience in distance learning, the University faces ongoing security challenges—and nowhere is that more serious than in the administration and evaluation of online exams. FFHS has long provided some exams online, and the requirement to secure data while preventing cheating has always been important. In recent years, however, the need for comprehensive security became even more pressing as the COVID pandemic increased the demand for entirely remote learning. In addition, the complexity of online computing has also increased rapidly with advances in technologies such as artificial intelligence (AI), as well as the increasingly sophisticated threats to data integrity posed by criminal hacking and fraud. To help meet the challenge of securing all aspects of the online exam process, FFHS joined forces with Intel Corporation to bolster the strength of their security capabilities in AI-based identity checking, fraud detection, and more. Implementing Secure Exams In preparation for this project, FFHS evaluated their previous examination system for two years. During that time, they identified a number of needs that could be met by a “virtual proctor” that would help detect and deal with fraud and cheating during exams. These needs included: 1. Significantly reduce the manual effort and hours required to monitor remote exams and to handle the massive amount of audio and video produced (more than 6.5 terabytes per semester). 2. Tie new functionality to the existing exam support system. 3. Protect exam data and students’ privacy during remote examinations, complying with privacy and security laws like the EU General Data Protection Regulation (GDPR). The initial approach to developing a digital system enabling secure remote exams was to implement an asynchronous analysis and evaluation process capable of recording video of students taking the exam. Because previous remote exams had demanded a significant effort by FFHS staff to monitor and evaluate, the new system would be designed to significantly reduce the human labor required. Swiss Distance University of Applied Sciences, Brig, Switzerland The Swiss Distance University of Applied Sciences (FFHS) delivers remote, online educational programs in disciplines such as IT, Engineering, Business, Communications, and more. Courses are tailored for continuing learners and learners who are unable to attend traditional classes on campus. Students can access courses on their own schedule, from anywhere in the world where connectivity is available. Fully accredited by the Swiss Accreditation Council, FFHS is a true e-university that provides all resources online, including course materials, virtual classrooms, interactive tools, and exams for bachelor’s and master’s degree programs. Having provided flexible, high-quality education for over 25 years, FFHS programs are well-respected and its graduates successful, with over 93% employed in work that matches their academic preparation. *Source: Erudera.com, https://erudera.com/ news/93-of-swiss-higher-education-graduates-get- employed-in-their-field-of-study-official-data-shows Solution Brief | Security in Education To help achieve this, FFHS migrated the exam workload and models to Intel® DevCloud for the Edge, a cloud-based development platform that enables developers to create and test edge computing applications quickly. FFHS began with student-identity-detection they had built with client API code in OpenCV and Python that had its own FaceID detection framework. Intel engineers worked with FFHS to transform
