A Comprehensive Survey on AI-Powered Personalized CAPTCHA Systems Leveraging User Preferences and Behavior

Abstract

This research paper aims to present a detailed survey of recent advancements in CAPTCHA systems utilizing Artificial Intelligence (AI) to generate personalized, image-based and text based CAPTCHA puzzles for the user’s. By analyzing the user’s data such as preferences, browsing history, and behavioral data, these CAPTCHA systems learn the individual users while enhancing security. The system also includes time-sensitive constraints to counter third-party bot attacks. The paper dives into the evolution of CAPTCHA systems, focusing on AI-based CAPTCHA generation, image-based CAPTCHAs, and methods used by bots to crack these security measures. We also go through many technical algorithms and methods such as behavior analysis, GANs, CNNs, and reinforcement learning that make CAPTCHAs more secure and adaptive.