Secure Deployment of Generative AI in Cloud Environments
DOI:
https://doi.org/10.14738/tecs.1403.10456Keywords:
AI, Generative AI, Cloud Security, Large Language Models, Machine LearningAbstract
Generative Artificial Intelligence (GenAI) models have become widely adopted through cloud computing platforms such as AWS, Microsoft Azure, and Google Cloud. Models such as ChatGPT and Gemini are transforming industries ranging from education and healthcare to enterprises and public services. Cloud environments provide scalability, cost efficiency, and ease of deployment; however, they also introduce complex privacy and security challenges. GenAI models are susceptible to sophisticated attacks such as prompt injection, model inversion, unauthorized access through insecure APIs, and data leakage. This paper examines security and privacy risks in cloud-hosted GenAI systems across data, training, deployment, and interface stages. It reviews mitigations like AI firewalls, differential privacy, and secure enclaves, and explores secure and trustworthy GenAI deployments.
References
T. Brown et al. (2020). Language models are few-shot learners. Advances in Neural Information Processing Systems (NeurIPS 2020), vol. 33, pp. 1877–1901.
N. Carlini et al. (2023). Extracting training data from diffusion models. Proc. 32nd USENIX Security Symposium (USENIX Security ’23), pp. 5253–5270.
OpenAI, (2022). Introducing ChatGPT. Retrieved from https://openai.com/blog/chatgpt
A. Qayyum et al. (2020). Securing machine learning in the cloud: A systematic review of cloud machine learning security. Frontiers in Big Data, vol. 3, 2020, Art. no. 587139. Scholar Publishing
Cloud Security Alliance, (2024). Top Threats to Cloud Computing 2024. Retrieved from https://cloudsecurityalliance.org/artifacts/top-threats-to-cloud-computing-post-pandemiceleven-survey-report
A. Luqman et al. (2024). Privacy and security implications of cloud-based AI services: A survey. arXiv preprint arXiv:2402.00896.
F. Perez and I. Ribeiro, (2022). Ignore previous prompt: Attack techniques for language models. arXiv preprint arXiv:2211.09527.
K. Zhao et al., (2025). A survey on model extraction attacks and defenses for large language models. arXiv preprint arXiv:2506.22521.
S. Ahmadi, (2023). Next generation AI-based firewalls: A comparative study. International Journal of Computer (IJC), vol. 49, no. 1, pp. 245–262.
M. Abadi et al. (2016). Deep learning with differential privacy. Proc. ACM SIGSAC Conf. Computer and Communications Security (CCS ’16), Vienna, Austria, 2016, pp. 308–318.
K. Bonawitz et al. (2019). Towards federated learning at scale: System design. Proc. Machine Learning and Systems (MLSys), 2019, pp. 374–388.
Microsoft, (2023). Confidential AI. Microsoft Learn, 2023. Retrieved from https://learn.microsoft.com/en-us/azure/confidential-computing/confidential-ai
G. Chen et al. (2018). SGXpectre attacks: Stealing Intel secrets from SGX enclaves via speculative execution. arXiv preprint arXiv:1802.09085.
National Institute of Standards and Technology, (2023). Artificial intelligence risk management framework (AI RMF 1.0). NIST AI 100-1, 2023. Retrieved from https://www.nist.gov/publications/artificial-intelligence-risk-management-framework-airmf-10
European Parliament and Council of the European Union, (2024). Artificial Intelligence Act. Retrieved from https://artificialintelligenceact.eu/
Microsoft Research, (2022). Powering the next generation of trustworthy AI in a confidential cloud using NVIDIA GPUs. Retrieved from https://www.microsoft.com/enus/research/blog/powering-the-next-generation-of-trustworthy-ai-in-a-confidential-cloudusing-nvidia-gpus/
Microsoft, (2025). Protect AI assets from emerging threats and vulnerabilities using Microsoft Defender. Microsoft Learn, 2025. Retrieved from https://learn.microsoft.com/enus/defender-xdr/security-for-ai/defender-security-for-ai
Amazon Web Services, (2024). Layer 3: Security and governance for generative AI platforms on AWS. AWS Prescriptive Guidance, 2024. Retrieved from https://docs.aws.amazon.com/prescriptive-guidance/latest/strategy-enterprise-ready-genai-platform/security.html
Google Cloud, (2025). Well-Architected Framework: Security, privacy, and compliance pillar. Retrieved from https://cloud.google.com/architecture/framework/security Scholar Publishing
P. Kairouz et al. (2021). Advances and open problems in federated learning. Foundations and Trends in Machine Learning. vol. 14, no. 1–2, pp. 1–210.
OWASP Foundation, (2024). Top 10 risks for large language model applications. Retrieved from https://owasp.org
I. Goodfellow, J. Shlens, and C. Szegedy, (2014). Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572.
N. Papernot et al., (2017). Practical black-box attacks against machine learning. Proc. ACM Asia Conf. Computer and Communications Security (AsiaCCS), 2017, pp. 506–519.
Akamai Technologies. Firewall for AI. Retrieved from https://www.akamai.com/products/firewall-for-ai
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Transactions on Engineering and Computing Sciences
This work is licensed under a Creative Commons Attribution 4.0 International License.
