The University of Michigan-Flint announced a groundbreaking partnership with NetSTAR, a global leader in web filtering and Internet security solutions. The collaboration aims to develop advanced artificial intelligence technologies to improve internet safety, particularly for vulnerable populations like children and the elderly. The project will also boost productivity by blocking access to harmful content, protecting sensitive information and minimizing distractions in educational and professional settings.
The partnership was officially announced during a signing ceremony at the Symposium on Cybersecurity Horizons: Unveiling Pathways, Addressing Challenges, and Shaping Research and Innovation, the first in a series of annual conferences hosted by UM-Flint. Center for Cybersecurity.
Nearly 200 attendees gathered at the university’s Riverfront Conference Center to see presenters from Microsoft, Amazon, General Motors and many other institutions, each sharing insights on career paths and industry challenges.
Khalid Malik, professor of computer science and director of the UM-Flint Cybersecurity Center, coordinated the event and welcomed NetSTAR executives, including Yasuaki Arakawa, the company’s chief technology officer, who has worked with Malik in the past.
“I first met Dr. Malik around 2010, when we were both working at NetSTAR in Tokyo,” Arakawa said. “We have collaborated on malware detection and website analysis, which has been very successful. Our shared experience was a key factor in NetSTAR’s decision to work with UM-Flint.”
After Malik moved to the United States, the two stayed in touch via LinkedIn, keeping the door open for future partnerships in cybersecurity. Finally, in May 2024, Malik and Arakawa began a renewed collaboration aimed at making the Internet safer for users, especially minors.
Their joint work focuses on AI technologies capable of detecting and categorizing malicious websites in real time. The traditional method for eliminating dangerous URLs is to manually update databases. But with millions of web addresses created and changed every day, the future of cybersecurity depends on AI technologies that can automatically detect harmful content like pornography or phishing sites.
According to Malik, the main obstacles to development include encrypted DNS queries, real-time categorization, changing website content and the multimodal nature of the Internet.
“In the future, many people will use the Internet with encrypted DNS queries, which hide their activity from ISPs,” he said. “Encrypted traffic makes it difficult to categorize and filter harmful websites. We need to categorize URLs in real time and cannot always trust previously assigned categories.”
Neuro-symbolic AI could provide a solution by enabling more sophisticated analysis and decision-making. The process contrasts with traditional deep learning, which, in Malik’s experience, is excellent at pattern recognition but falls short at reasoning.
“Our ultimate goal is to strike a balance between Internet security and privacy,” Malik said. “To meet these challenges, we are developing neuro-symbolic techniques for URL classification, which involve a combination of symbolic reasoning and multimodal deep AI capable of processing audio, video and text.
An AI model capable of accurately assessing online threats would represent a technological leap forward in cybersecurity, and in his remarks at the Cybersecurity Horizons Symposium, Arakawa highlighted the importance of success of the project.
“Our goal is to use technology to create solutions that improve the safety and security of the Internet for everyone,” Arakawa said. “But this collaboration is more than just a technology project: it’s about making a real difference in the world. By combining the strengths of NetSTAR with the exceptional skills of UM-Flint, we aim to address the complex cybersecurity challenges we face today.
To learn more, visit the UM-Flint Center for Cybersecurity Web page.
