Crossref DOIMemberCiteFactoropen accessissn
SJIF Impact Factor 8.659 | ISSN: 2582-3930
IJSREM Logo

International Journal of Scientific Research in Engineering and Management

(Gateway to Global Scholars - An International, Scholarly, Multi-Disciplinary, Open Access Journal)

Peer-Reviewed Journal Designed to Meet UGC's Quality Parameters

HomeCurrent IssuePast IssuesPay OnlineConferencesEditorial BoardAbout UsContact Us
IJSREM

Loading article...

Follow Us

  • Google Scholar
  • ResearchGate
  • Academia.edu
  • Mendeley
  • LinkedIn
  • Facebook
  • Instagram

Journal Policies

  • Peer Review Policy
  • Publication Ethics
  • Copyright Infringement
  • Refund & Cancellation Policy
  • Terms and Conditions
  • Privacy Policy

Journal Identifiers & Licenses

  • ISSN Verified: 2582-3930ISSN Verified: 2582-3930
  • Crossref DOI: 10.55041Crossref DOI: 10.55041
  • SJIF Ranking: 08.659SJIF Ranking: 08.659
  • Licensed under CC BY 4.0Licensed under CC BY 4.0
  • ISO Certification: 9001:2015ISO Certification: 9001:2015
  • Indexed & Cited In: Google ScholarIndexed & Cited In: Google Scholar

Contact Us

International Journal of Scientific Research in Engineering and Management (IJSREM)

#62/1, New No 7, 1st Cross, 2nd Main, Ganganagar, R T Nagar, Bangalore North, Bangalore, Karnataka, India – 560032
[email protected]+91 93911 67991
logos
logos
Working Hours

Mon – Sat: 9:00 AM – 6:00 PM

Sunday: 10:00 AM – 4:00 PM

+91 93911 67991

© 2026 IJSREM – International Journal of Scientific Research in Engineering and Management. This is an open-access journal. Articles are published under the Creative Commons Attribution (CC BY 4.0) License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

DisclaimerPrivacy PolicyTerms and Conditions
ArticlePeer ReviewedOpen AccessPlagiarism Passed
Paper ID:IJSREM69156

Defending AI-Assisted Assessments Against Adversarial Prompt Injection: A Multi-Layered Architecture With Real-Time Exam Termination

Stream/Area:

-

Subject:

-

Paper ID:

IJSREM69156

Volume:

10 |

11 |

Nov 2026

Pages:

1-6

Publication Frequency:

Monthly (12 issues per year)

DOI
Digital Object Identifier
10.55041/IJSREM69156

License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Share this article:

Authors

Isha Mishra

Isha Mishra

Department of Computer Science

Shri Shankaracharya Technical Campus

Download Author Certificate
Meenakshi Tuwani

Meenakshi Tuwani

Department of Computer Science

Shri Shankaracharya Technical Campus

Download Co-Author Certificate
Isha Gupta

Isha Gupta

Department of Computer Science

Shri Shankaracharya Technical Campus

Download Co-Author Certificate
Astha Sharma

Astha Sharma

Department of Computer Science

Shri Shankaracharya Technical Campus

Download Co-Author Certificate

Mentors

Prof. Durgeshwari Sahu

Prof. Durgeshwari Sahu

Department of Computer Science

Shri Shankaracharya Technical Campus

Download Mentor Certificate

Author Services & Downloads

Full ArticlesAcceptance Certificate

Abstract

The rapid adoption of artificial intelligence (AI) in online assessments has introduced critical vulnerabilities, particularly adversarial prompt injection attacks intended to manipulate automated grading and proctoring models. This paper presents a secure AI-based examination framework that combines five foundational defense mechanisms with an Adversarial Prompt Injection Detection and Immediate Termination (APID-IT) protocol. Federated learning is used to support privacy-preserving proctoring, while deep learning supports behavioral and speech anomaly detection. Assessment integrity is strengthened through decentralized exam-paper encryption, AI-assisted question randomization, blockchain-backed audit logs, and compute isolation. At the core of the proposed system, an input-sanitization and semantic-intent classification pipeline monitors candidate inputs before they reach the grading language model. In the simulated evaluation described in this paper, 500 synthetic candidate sessions were considered, including 150 adversarial sessions. The reported results include a 99.2% prompt-injection detection rate, a 0.4% false-positive termination rate, and an average termination latency of 118 ms. These figures are simulation results and should be interpreted in the context of the stated experimental setup.

KEYWORDS

AI-assisted assessment prompt injection federated learning blockchain multimodal anomaly detection trusted execution environment examination security real-time termination.

Cite This Article

Isha Mishra, Meenakshi Tuwani, Isha Gupta, Astha Sharma, Prof. Durgeshwari Sahu, "Defending AI-Assisted Assessments Against Adversarial Prompt Injection: A Multi-Layered Architecture With Real-Time Exam Termination," International Journal of Scientific Research in Engineering and Management, vol. 10, no. 11, pp. 1-6, Nov 2026. https://doi.org/10.55041/IJSREM69156

Full Article/Conclusion

This paper presents a unified AI-based examination security architecture combining federated learning, blockchain auditability, decentralized encryption, multimodal anomaly detection, and compute isolation with an active Adversarial Prompt Injection Detection and Immediate Termination mechanism. The principal contribution is the proposed cognitive-layer defense that screens candidate inputs before they reach an AI grader and links high-confidence incidents to an auditable termination process. The source paper reports a 99.2% detection rate, 118 ms termination latency, and 0.4% false-positive termination rate in a simulated evaluation. Further work should validate these results through reproducible experiments and real-world pilot deployments, with particular attention to false positives, fairness, privacy, and resilience against adaptive attacks.

References

[1]           H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A.             y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Proc. 20th Int. Conf. Artificial Intelligence and Statistics (AISTATS), 2017, pp. 1273–1282. [Online]. Available: https://arxiv.org/abs/1602.05629 [2] P. Kairouz et al., “Advances and open problems in federated learning,” Foundations and Trends in Machine Learning, vol. 14, nos. 1–2, pp. 1–210, 2021. [Online]. Available: https://arxiv.org/abs/1912.04977 [3]           E. Androulaki et al., “Hyperledger Fabric: A distributed operating system for permissioned blockchains,” in Proc. 13th EuroSys Conf. (EuroSys ’18), 2018, pp. 1–15. [Online]. Available: https://arxiv.org/abs/1801.10228 [4] M. Turkanović, M. Hölbl, K. Košič, M. Heričko, and A. Kamišalić, “EduCTX: A blockchain-based higher education credit platform,” IEEE Access, vol. 6, pp. 5112–5127, 2018. [Online]. Available: https://ieeexplore.ieee.org/document/82471667 [5] A. Shamir, “How to share a secret,” Communications of the ACM, vol. 22, no. 11, pp. 612–613, 1979. [Online]. Available: https://dl.acm.org/doi/10.1145/359168.359176 [6] G. Kurdi, J. Leo, B. Parsia, U. Sattler, and A. Al-Emari, “A systematic review of automatic question generation for educational purposes,” International Journal of Artificial Intelligence in Education, vol. 30, no. 1, pp. 121–204, 2020. [Online]. Available: https://link.springer.com/article/10.1007/s40593-019-00186-y [7] Y. Atoum, L. Chen, A. X. Liu, S. D. H. Hsu, and X. Liu, “Automated online exam proctoring,” IEEE Transactions on Multimedia, vol. 19, no. 7, pp. 1609–1624, 2017. [Online]. Available: https://ieeexplore.ieee.org/document/7862766 [8] A. Radford et al., “Robust speech recognition via large-scale weak supervision,” in Proc. 40th Int. Conf. Machine Learning (ICML), 2023, pp. 28492–28518. [Online]. Available: https://arxiv.org/abs/2212.04356 [9] V. Costan and S. Devadas, “Intel SGX explained,” IACR Cryptology ePrint Archive, Paper 2016/086, 2016. [Online]. Available: https://eprint.iacr.org/2016/086 [10]        B. Jacob et al., “Quantization and training of neural networks for efficient integer-arithmetic-only inference,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2018, pp. 2704–2713. [Online]. Available: https://arxiv.org/abs/1712.05877

Editorial & Review Transparency

Publication Date

11/18/2026

Peer Review Type

Double-Blind Peer Review

Review Rounds: 2

Documents

Review Report (Confidential)
Copyright Form (Office)

Article Metrics & Impact

Article Views

0

Total views since publication

PDF Downloads

3

Total downloads since publication

Geographic Reach

5

Indexing, Archiving & Preservation

Google Scholar
Academia
ResearchGate
Crossref
SSRN
Semantic Scholar
Thomson Reuters
WorldCat
Mendeley
OpenAlex
EBSCO
Zenodo

Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.

Ethics, Compliance & Transparency

Plagiarism Screening

This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.

Conflict of Interest

The authors declare no conflict of interest related to this research work.

Funding Information

No funding information available for this research.

Similar & Recommended Articles

Similar Articles by Topic

Related research in the same subject area

Loading similar articles...