Artificial Intelligence (AI) is increasingly prevalent in our world—from obvious examples in the home like smart doorbells and Alexa to innovative uses like optimizing an Uber route or helping viewers select the best Netflix options. But with AI seeping into every aspect of life comes serious cybersecurity issues and the threat of attackers targeting AI algorithms. Recommending an uninteresting movie might not sound like a big deal but imagine a self-driving car failing to recognize a stop sign or someone receiving the wrong medical treatment because of a vulnerability in an AI system.
Traditionally, AI and cybersecurity are separate topics of study in a university setting. Casey Fiesler and Robin Burke, faculty members of the Department of Information Science (INFO), and security expert Eric Wustrow of the Department of Electrical, Computer and Energy Engineering (ECEE), were recently awarded a $298 thousand grant from the National Science Foundation (NSF) to create new coursework specifically addressing this knowledge gap.
“We’re really fortunate to have the support of instructors across CMCI and Engineering who teach AI-related courses and excited to both work with a team of students on this research and then to integrate the curriculum we develop into their courses,” says Fiesler.
Fiesler’s team will build the curriculum in collaboration with those instructors and will rely on a robust evaluation effort heavily involving students to ensure learning goals and efficacy are met.
“We’ve already been doing work to integrate ethics into programming classes as part of the Responsible Computer Science Challenge and the response from students has been really positive,” Fiesler says. “Security is another topic that makes a lot of sense to be cross-cutting, particularly for contexts like AI.”
Once complete, the curriculum will be broadly distributed to other educators, and the educational strategies the team designs will also provide insight into how AI and cybersecurity experts can collaborate in the field today.
This work is funded by the National Science Foundation (NSF), Division of Graduate Education in the program of Secure and Trustworthy Cyberspace, Award # 2115028.