Ananya R. Bhat,
Srushti Ravindra Chigadani,
Yathin G. Kummar,
Rashmi T.V,
- Student, Department of ISE, BNM Institute of Technology, Bengaluru, India
- Student, Department of ISE, BNM Institute of Technology, Bengaluru, India
- Student, Department of ISE, BNM Institute of Technology, Bengaluru, India
- Assistant Professor, Department of ISE, BNM Institute of Technology, Bengaluru, India
Abstract
CAPTCHA, which stands for Completely Automated Public Turing test to tell Computers and Humans Apart, is a tool that separates human users from automated bots. It is commonly used in web applications to stop bulk registrations, spam submissions, credential stuffing, and misuse of online services. Typically, CAPTCHAs involve recognizing distorted text, selecting images, or transcribing audio. These tasks are easy for people but challenging for automated programs. However, recent advances in artificial intelligence have made these methods less effective. Machine learning models and multimodal AI systems can now solve text, image, and audio CAPTCHAs with nearly human-like or even better accuracy. This poses serious risks, such as creating fake accounts, ticket scalping, and large-scale automated misuse of websites. As a result, researchers are looking into new methods, like video-based CAPTCHAs. These use short dynamic clips that introduce motion or timing challenges, relying on temporal reasoning. Such tasks are straightforward for humans but still difficult for AI, which often has trouble tracking changes across frames or recognizing complex motion boundaries. This survey and analysis show that static CAPTCHAs are no longer enough in the age of AI. Video-based CAPTCHAs, which include replay protection through hashing and timestamping along with adversarial testing to counter new attacks, provide a more practical and sustainable solution for modern web security.
Keywords: Adversarial testing, AI-resistant security, bot detection, temporal reasoning, video CAPTCHA, web application security
[This article belongs to Journal of Web Engineering & Technology ]
References
- Kopp M, Nikl M, Holena M. Breaking CAPTCHAs with Convolutional Neural Networks. InITAT 2017 Sep (pp. 93-99).
- Guerar M, Verderame L, Migliardi M, Palmieri F, Merlo A. Gotta CAPTCHA’Em all: a survey of 20 Years of the human-or-computer Dilemma. ACM Computing Surveys (CSUR). 2021 Oct 7;54(9):1-33.
- Yusuf MO, Srivastava D, Singh D, Rathor VS. Multiview deep learning-based attack to break text-CAPTCHAs. International journal of machine learning and cybernetics. 2023 Mar;14(3):959-72.
- Aubry P, Devoivre J, Carron D, Fernandez S, Duda A, Korczyński M. Bypassing Audio reCAPTCHA with Automatic Speech Recognition Models. 2025 IEEE European Symposium on Security and Privacy Workshops (EuroS&PW) 2025 Jun 30. pp. 1-5.
- Johri E, Dharod L, Joshi R, Kulkarni S, Kundle V. Video Captcha Proposition based on VQA, NLP, Deep Learning and Computer Vision. In2022 5th International Conference on Advances in Science and Technology (ICAST) 2022 Dec 2. pp. 196-200.
- Lin L, Wang K, Abdallah M, Cai W. BounTCHA: A CAPTCHA Utilizing Boundary Identification in Guided Generative AI-extended Videos. arXiv preprint arXiv:2501.18565. 2025 Jan 30.
- Patil N, Patil R. Achieving flatness: with video captcha, location tracking, selecting the honeywords. In2018 International Conference on Smart City and Emerging Technology (ICSCET) 2018 Jan 5. pp. 1-6.
- Wu Z, Xue Y, Feng Y, Wang X, Song Y. MCA-bench: A multimodal benchmark for evaluating CAPTCHA robustness against VLM-based attacks. In Proceedings of the AAAI Conference on Artificial Intelligence 2026 Mar 14 (Vol. 40, No. 44, pp. 38039-38047).
- Gurale M, Borate K, Sangitrao RR, Tiwari G. Enhanced Security Through Honeywords and Video CAPTCHA: An Innovative Authentication and Tracking Mechanism. In2025 Fifth International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT) 2025 Jan 9. pp. 1-6.
- Moradi M, Moradi M, Palazzo S, Rundo F, Spampinato C. Image CAPTCHAs: When deep learning breaks the mold. IEEE Access. 2024 Aug 13;12:112211-31.
Journal of Web Engineering & Technology
| Volume | 13 | |
| Issue | 02 | |
| Received | 03/02/2026 | |
| Accepted | 12/03/2026 | |
| Published | 10/04/2026 | |
| Publication Time | 66 Days |