Tamper-Proof Privacy Auditing for Artificial Intelligence Systems
Open accessPublished version of a preprint
Abstract
Privacy audit logs are used to capture the actions of participants in a data sharing environment in order for auditors to check compliance with privacy policies. However, collusion may occur between the auditors and participants to obfuscate actions that should be recorded in the audit logs. In this paper, we propose a Linked Data based method of utilizing blockchain technology to create tamper-proof audit logs that provide proof of log manipulation and non-repudiation.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
Tamper-evident logs and immutable agent identifiers can support attribution and enable identification of suspicious behavioural patterns across agent populations.
“Complementary mechanisms such as tamper-evident logs (Sutton & Samavi, 2018) and immutable agent identifiers (Chan et al., 2024b) can support attribution and enable the identification of suspicious behavioural patterns across agent populations (Ju et al., 2024).”
From Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents (2025) · cites this paper as Sutton & Samavi, 2018
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 15
- Accorsi, 2009] Rafael Accorsi. Log data as digital evidence: What secure logging protocols have to offer? In Computer Software and Applications Conference, 2009. COMP- SAC'09. 33rd Annual IEEE International, volume 2, pages 398-403. IEEE, 2009.
- Denis Butin, Marcos Chicote, and Daniel Le Métayer. Log design for accountability. IEEE Security and Privacy Workshops, 2013.
- Jeremy J Carroll, Christian Bizer, Pat Hayes, and Patrick Stickler. Named graphs. Web Seman- tics: Science, Services and Agents on the World Wide Web, 3(4):247-267, 2005.
- Cucurull and Puiggalí, 2016] Jordi Cucurull and Jordi Puig- galí. Distributed immutabilization of secure logs. In In- ternational Workshop on Security and Trust Management, pages 122-137. Springer, 2016.
- Heath and Bizer, 2011] Tom Heath and Christian Bizer. Linked data: Evolving the web into a global data space. Synthesis lectures on the semantic web: theory and tech- nology, 1(1):1-136, 2011.
- Kasten et al., 2014] Andreas Kasten, Ansgar Scherp, and Peter Schauß. A framework for iterative signing of graph data on the web. In European Semantic Web Conference, pages 146-160. Springer, 2014.
- Kasten, 2016] Andreas Kasten. Secure semantic web data management: confidentiality, integrity, and compliant availability in open and distributed networks. PhD thesis, University of Koblenz and Landau, Germany, 2016.
- Florian Kleedorfer, Yana Panchenko, Christina Maria Busch, and Christian Huemer. Verifiability and traceability in a linked data based messaging system. In Proceedings of the 12th International Conference on Semantic Systems, pages 97-100. ACM, 2016.
- Manu, 2017] Sporny Manu. Building better blockchains. In Linked Data in Distributed Ledgers Workshop Keynote. WWW2017, 2017.
- Ralph C Merkle. Protocols for public key cryptosystems. In Security and Privacy, 1980 IEEE Sym- posium on, pages 122-122. IEEE, 1980.
- Reza Samavi and Mariano P Consens. Publishing privacy logs to facilitate transparency and accountability. Journal of Web Semantics, 2018.
- Vassilios Stathopoulos, Panayio- tis Kotzanikolaou, and Emmanouil Magkos. Secure log management for privacy assurance in electronic commu- nications. computers & security, 27(7-8):298-308, 2008.
- Vipin Swarup, Len Seligman, and Arnon Rosenthal. A data sharing agreement framework. In International Conference on Information Systems Secu- rity, pages 22-36. Springer, 2006.
- Yue Tong, Jinyuan Sun, Sherman SM Chow, and Pan Li. Cloud-assisted mobile-access of health data with privacy and auditability. IEEE Journal of biomedical and health Informatics, 18(2):419-429, 2014.
- Daniel J Weitzner, Harold Abelson, Tim Berners-Lee, Joan Feigenbaum, James Hendler, and Gerald Jay Sussman. Information accountability. Commu- nications of the ACM, 51(6):82-87, 2008.
Cited by
Sorted by citations
Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents
de Witt, C. S., Krawiecka, K., Krawczuk, I., Hagag, B., Anderson, W. L., Belcak, P., Bucknall, B., Cai, X., Chopra, A., Cohen, D., et al. (2025) · arXiv · doi:10.48550/arxiv.2505.02077 · Cited by 2 · see citing works
“Similarly, tamper-evident logs (Sutton & Samavi, 2018) and immutable agent identifiers (Chan et al., 2024b) could be used to detect suspicious patterns among networks of agents (Ju et al., 2024) and allow for faster remediation.”
- Authors
- Sutton, A., & Samavi, R.
- Affiliations
- McMaster University (CA)
- ORCID
- Samavi, Reza
- Venue
- Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence
- Published
- 2018-07
- Pages
- 5374-5378
- OpenAlex
- W2834398869
- Cited by
- 3
- References
- 0
- Citations by year
- 2023: 1 · 2022: 1 · 2021: 1
- FWCI
- 0.20
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- International Joint Conference on Artificial Intelligence (publishedVersion)
- Funders
- Natural Sciences and Engineering Research Council of Canada
- Topics
- Physical Sciences › Computer Science › Information Systems › Blockchain Technology Applications and Security
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Information Systems › Cloud Data Security Solutions
- Keywords
- privacy auditing, blockchain, linked data, collusion, non-repudiation, data sharing, privacy policy compliance, artificial intelligence systems
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
