Privacy and Data Protection in ChatGPT and Other AI Chatbots: Strategies for Securing User Information
Open accessThe DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index.
Abstract
The evolution of artificial intelligence (AI) and machine learning (ML) has led to the development of sophisticated large language models (LLMs) that are used extensively in applications such as chatbots. This research investigates the critical issues of data protection and privacy enhancement in the context of LLM-based chatbots, with a focus on OpenAI's ChatGPT. It explores the dual challenges of safeguarding sensitive user information while ensuring the efficiency of machine learning models. It assesses existing Privacy-Enhancing Technologies (PETs) and proposes innovative methods, such as differential privacy, federated learning, and data minimization techniques. The study also includes a survey of Chatbot users to measure their concerns related to Data Privacy with the use of these LLM-based applications. This study is meant to serve as a comprehensive guide for developers, policymakers, and researchers, contributing to the discourse on data protection in artificial intelligence.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Severity of personal data leaks: Concerns about personal data leaks are paramount79.”
From On opportunities and challenges of large multimodal foundation models in education (2025) · cites this paper as [79]
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Exploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,887
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381 · cited 2×
“Verifying and securely sourcing data can ensure the integrity and legality of the data, but it may also limit the amount of data available for training, possibly affecting the performance and usability of the model (Goodfellow, I., et al., 2015).”
k-ANONYMITY: A MODEL FOR PROTECTING PRIVACY
SWEENEY, L. (2002) · International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems · doi:10.1142/s0218488502001648 · Cited by 8,571
Calibrating Noise to Sensitivity in Private Data Analysis
Dwork, C., McSherry, F., Nissim, K., & Smith, A. (2006) · Lecture Notes in Computer Science · doi:10.1007/11681878_14 · Cited by 7,272
Membership Inference Attacks Against Machine Learning Models
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.41 · Cited by 4,754 · cited 3×
“The impact on performance would likely be minimal unless the system becomes overloaded with automated queries (Biggio, B., & Roli, F.,2018) (Shokri, R., 2017) (Akhawe, D., 2013) (Xu, Minrui, et al., 2023) undergo regular Cybersecurity audits, there is a clear consensus for AI systems to adhere to data protection regul…”
How to generate and exchange secrets
Yao, A. C.-C. (1986) · 27th Annual Symposium on Foundations of Computer Science (sfcs 1986) · doi:10.1109/sfcs.1986.25 · Cited by 3,733
Differential Privacy: A Survey of Results
Dwork, C. (2008) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-79228-4_1 · Cited by 3,542 · cited 4×
“If successful, the attacker could use the extracted model for malicious purposes, potentially undermining the security and integrity of the original system (Li, Haoran, et al., 2023) (Tramèr, F., et al., 2023) (Shokri, R., et al., 2017) (Dwork, C. et al., 2006). v) Data Poisoning: This is a type of attack where the at…”
InfluentialArtificial hallucinations in ChatGPT: implications in scientific writing
Alkaissi, H., & McFarlane, S. I. (2023) · Cureus · doi:10.7759/cureus.35179 · Cited by 1,540
Research works printed but not linked: 13
- Openai, 2021, ChatGPT: Your AI Friend
- Openai, 2021, OpenAI Data Usage Policy
- Openai, 2021, Fine-Tuning Large Language Models: Dataset and Safety. OpenAI Blog
- Ou Zheng, 2023, ChatGPT is on the horizon: Could a large language model be all we need for Intelligent Transportation?
- Danya Sherbini, 2023, Has technological innovation lost the plot? An interview with AI ethicist Dr. Shannon Vallor, Chicago Policy Review (Online)
- Walid Hariri, 2023, Unlocking the Potential of ChatGPT: A Comprehensive Exploration of its Applications, Advantages, Limitations, and Future Directions in Natural Language Processing
- Karen Renaud, 2023, From ChatGPT to HackGPT: Meeting the Cybersecurity Threat of Generative AI, 64, 1
- G Cormode, 2019, Data Anonymization Revisited, SIGMOD Rec, 41, 6
- C Dwork, 2006, Calibrating Noise to Sensitivity in Private Data Analysis, Theory of Cryptography Conference
- N Papernot, 2016, Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data, ICLR
- I Goodfellow, 2015, Explaining and Harnessing Adversarial Examples, ICLR
- R Shokri, 2017, Membership inference attacks against machine learning models, 2017 IEEE Symposium on Security and Privacy, 3, 10.1109/SP.2017.41 · DOI · Paper page
- D Akhawe, 2013, Here's my cert, so trust me
Cited by
A survey on large language model (LLM) security and privacy: The Good, The Bad, and The Ugly
Yao, Y., Duan, J., Xu, K., Cai, Y., Sun, Z., & Zhang, Y. (2024) · High-Confidence Computing · doi:10.1016/j.hcc.2024.100211 · Cited by 1,083
Casheekar, A., Lahiri, A., Rath, K., Prabhakar, K. S., & Srinivasan, K. (2024) · Computer Science Review · doi:10.1016/j.cosrev.2024.100632 · Cited by 211
Privacy-Preserving Techniques in Generative AI and Large Language Models: A Narrative Review
Feretzakis, G., Papaspyridis, K., Gkoulalas-Divanis, A., & Verykios, V. S. (2024) · Information · doi:10.3390/info15110697 · Cited by 145
“To strengthen data anonymization, advanced techniques such as k-anonymity, l-diversity, and t-closeness are employed to prevent re-identification attacks.”
InfluentialChatGPT Promises and Challenges in Education: Computational and Ethical Perspectives
Adel, A., Ahsan, A., & Davison, C. (2024) · Education Sciences · doi:10.3390/educsci14080814 · Cited by 108
“This may involve incorporating elements of self-determination theory, such as autonomy, competence, and relatedness, into the design and implementation of AI-driven educational experiences.”
Background
Metadata
- Authors
- Sebastian, G.
- Affiliations
- Georgia Institute of Technology (US)
- Corresponding author
- Sebastian, Glorin
- ORCID
- Sebastian, Glorin
- Venue
- SSRN Electronic Journal
- Publisher
- Elsevier BV
- Published
- 2023
- ISSN
- 1556-5068 (ISSN-L)
- OpenAlex
- W6959719920
- Cited by
- 56
- References
- 25
- Citations by year
- 2026: 2 · 2025: 10 · 2024: 32 · 2023: 12
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Locations
- SSRN Electronic Journal (acceptedVersion)
- Topics
- Social Sciences › Law › Digitalization, Law, and Regulation
- Social Sciences › Sociology and Political Science › Privacy, Security, and Data Protection
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Keywords
- privacy, data protection, ChatGPT, AI chatbots, personal data
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.