A new AI-driven chatbot developed by researchers at Georgia Tech and UDC is breaking new ground in cybersecurity. By analyzing social media sentiments, this innovative tool aims to predict and prevent cyberattacks before they happen.
Cyberattacks, once occasional disruptions, have grown as frequent and devastating as natural disasters. Highlighting the urgency of cybersecurity, a pioneering AI tool now demonstrates the capability to forecast these digital threats by analyzing social media.
Researchers from Georgia Institute of Technology’s Scheller College of Business and the University of the District of Columbia (UDC) have developed an innovative chatbot able to scrutinize social media sentiment and identify who might initiate the next significant cyberattack.
The chatbot, specifically designed to engage with users on platforms such as X (formerly Twitter), interacted with 100,000 users over a three-month period, engaging those who tweeted about news events or holidays, or retweeted cyberattack news. The critical component of the research involved performing sentiment analysis on the responses the chatbot received, evaluating users’ feelings, attitudes and moods to identify potential threats.
“AI can target a particular population to understand its expressions of approval, disapproval, or even intent to harm, attack, or misuse the technology,” John McIntyre, professor and founding director of the Georgia Tech Scheller College of Business Center for International Business Education and Research (CIBER), said in a news release.
While sentiment analysis on chatbots isn’t new, its application in cybersecurity is novel. Typically, global companies utilize chatbots to analyze customer reactions to brands and products. During the COVID-19 pandemic, chatbots were instrumental in gauging public sentiment toward vaccinations and preventive health measures. However, leveraging sentiment analysis to spot potential cybersecurity threats represents a groundbreaking and sophisticated endeavor.
The project, led by McIntyre together with UDC Associate Professors Amit Arora and Anshu Arora, uncovered how social media could become a resource for identifying cyber threats. This methodology is only the beginning, with potential to extend analysis across various languages and other digital platforms.
“As we move toward a world in which we’ll rely more and more on communication technologies and social media, there will be an increasing number of threats,” added McIntyre. “We must know how to counter such threats.”
Published in the journal Sustainability, this innovative research establishes a critical foundation in the ongoing battle against cyber threats.