Middle schoolers don’t usually spend their summer afternoons cracking secret codes.
But inside a computer lab in the School of Computing and Augmented Intelligence, part of the Ira A. Fulton Schools of Engineering at Arizona State University, groups of tweenagers gathered one day in May racing through password-cracking puzzles, deciphering secret messages and learning how hackers think.
The exercises were part of Desert CodeSprouts, an annual computer science workshop for middle school students hosted by Fulton Schools faculty members. The one-day camp introduces students to cybersecurity, robotics, artificial intelligence, or AI, and privacy through hands-on challenges designed to feel more like games than lectures.
For Ozgur Ozmen, a Fulton Schools assistant professor of computer science and engineering, the cybersecurity module wasn’t simply about entertaining students for an afternoon. It was about planting an idea.
“If we can teach kids about security early on, then in the future they’ll design technology with security in mind instead of treating it as an afterthought,” Ozmen says. “Hopefully, the next generation will build systems with fewer and fewer security problems.”
That approach reflects a recurring theme throughout Ozmen’s early career. He sees mentorship as a continuum, helping students discover cybersecurity in middle school, develop as researchers in college and eventually become mentors themselves.
The sooner, the better
Cybersecurity often suffers from a familiar problem. Products are designed to be fast, inexpensive and convenient. Security comes later, patched onto systems after vulnerabilities emerge.
Ozmen believes education follows a similar pattern. Too often, students discover cybersecurity only after they’ve already chosen a technical specialty. By then, many of the habits and assumptions behind system design are already established.
Desert CodeSprouts offers a different approach.
This year’s workshop introduced middle school students to cybersecurity through Caesar ciphers, password-cracking exercises and interactive challenges hosted on the pwn.college cybersecurity training platform. Rather than listening to lectures, students solved increasingly difficult problems while working alongside graduate student mentors.
The experience surprised Ozmen.
“Some students already had experience with lightweight coding. They were much more advanced than I expected them to be,” he says. “The new generation has definitely been raised with much more technology and designing educational modules for them in the future will be very challenging.”
Instead of intimidating him, that enthusiasm reinforced his belief that students are ready for deeper technical ideas much earlier than many educators assume.
Research begins with the right questions
For Ozmen, Desert CodeSprouts represents only the first step in a much longer journey. On campus, he mentors undergraduate researchers through ASU’s Fulton Undergraduate Research Initiative, or FURI.
Many arrive having never participated in research.
“They have a lot to give to research,” Ozmen says. “They have fresh minds, and it’s really fun for me to experience how they tackle their first research problems and become more independent decision-makers.”
His mentoring style reflects that philosophy.
Rather than assigning narrowly defined tasks, Ozmen asks students to immerse themselves in the latest cybersecurity literature, identify research papers that excite them and explain why. Only after discovering a student’s interests will he help them identify unanswered research questions.
The process gives students ownership over their work while teaching them how researchers think.
Those projects naturally grow out of Ozmen’s own research on securing emerging cyberphysical systems. His students are investigating problems such as machine learning defenses against attacks on disaster-response communication networks, secure communication between batteryless and battery-powered Internet of Things, or IoT, devices, and cybersecurity vulnerabilities in autonomous robots.
Together, they tackle real-world security challenges while giving students experience designing solutions that could shape future technologies.
Learning by building
Unlike research that exists primarily on paper, Ozmen’s students spend much of their time interacting with physical systems. His laboratory includes drones, robotic platforms, IoT devices and simulation environments that enable students to see cybersecurity vulnerabilities play out in real time.
Watching an autonomous vehicle crash inside a simulator or observing how compromised sensors behave makes cybersecurity tangible in a way that lectures alone cannot.
Those experiences also prepare students for careers beyond academia. Research projects require students to integrate programming, systems engineering and problem solving over months rather than weeks, producing work they can demonstrate to future employers.
Mentors making mentors
As Ozmen looks toward the future, he envisions something larger than a successful research laboratory.
He imagines a community through which middle school students become undergraduates, undergraduates become doctoral researchers and doctoral researchers become mentors themselves.
That process has already begun.
At Desert CodeSprouts, graduate students — not faculty members — led much of the cybersecurity instruction while Ozmen and others on the team guided the overall experience.
He hopes that model eventually becomes self-sustaining.
“I want my doctoral students to become great mentors,” Ozmen says. “When they develop those skills, they can create a culture in which students learn from one another.”
For Ozmen, cybersecurity isn’t only about protecting computers.
It’s about cultivating the people who will protect the technologies that haven’t been invented yet. Sometimes, that journey begins with a middle school student smiling after cracking a simple cipher for the very first time.
