Abstract
The rapid growth of web-based services has significantly increased the attack surface for malware distribution, particularly through non-secure or poorly protected websites. Traditional antivirus and endpoint protection systems are largely reactive, often detecting threats only after widespread infection has occurred. This delay allows emerging malware to compromise a large number of users before effective countermeasures are deployed.
The presented disclosure propose an AI-guided controlled web exposure framework designed to proactively identify emerging web-based malware and evaluate antivirus effectiveness under high-risk conditions. The proposed system operates within fully isolated virtual environments and utilizes multiple controlled administrator-level user accounts to simulate realistic privilege escalation scenarios. An artificial intelligence–driven exposure engine dynamically selects and prioritizes potentially risky websites based on adaptive risk assessment and feedback from prior infection outcomes.
The framework continuously monitors system behavior, file system activity, and network traffic to detect indicators of compromise and to assess antivirus response latency and effectiveness. By combining adaptive AI-driven exploration with controlled administrative privilege analysis, the proposed approach enables early detection of malicious web activity while maintaining strict ethical and legal safeguards. The proposal demonstrates the potential of proactive, intelligence-guided exposure systems to reduce the time gap between malware emergence and detection, thereby strengthening defensive cybersecurity mechanisms and improving overall user protection.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Niranjan, Chiranthan, "AI-Guided Controlled Web Exposure Framework for Early Detection of Web-Based Malware", Technical Disclosure Commons, (January 12, 2026)
https://www.tdcommons.org/dpubs_series/9177