Author ORCID Identifier
Anik Mukherjee: 0000-0001-9130-5720
Indranil Bose: 0000-0002-5737-966X
Abstract
CyberCom, recognized for its cloud-based security solutions that emphasize antivirus protection, malware detection, and endpoint security, has developed several key products: CyberComSECURE, CyberComLENS, and CyberComSHIELD. These products are designed to defend against system vulnerabilities and enable rapid responses to cyber threats. Central to the product line is a machine learning (ML) model trained on vast amounts of data related to malware and past cyber incidents, offering reliable detection of known malware. However, external security researchers discovered vulnerabilities within this model that rendered it ineffective in detecting mutated malware files. These altered files exploited the system's reliance on historical data, exposing the critical limitation of artificial intelligence (AI)-based cybersecurity tools in addressing new and evolving threats. The case describes what led to this scenario and raises the question how CyberCom should handle the crisis and future-proof its products against similar attacks.
Recommended Citation
Mukherjee, A., Bose, I., & Chatterjee, D. (In press). When the Bee Stings: CyberCom’s AI Vulnerability. Communications of the Association for Information Systems, 59, pp-pp. Retrieved from https://aisel.aisnet.org/cais/vol59/iss1/65
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
