# Privacy Washing: Detecting Internal Contradictions in Privacy Policies

arXiv: https://arxiv.org/abs/2609.02055 · HTML: https://mlsanity.com/paper/2609.02055/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 111 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 30 of 30)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing](https://mlsanity.com/paper/2111.09543/) (2021)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models](https://mlsanity.com/paper/2404.18796/) (2024)
- [ContractNLI: A Dataset for Document-level Natural Language Inference for Contracts](https://mlsanity.com/paper/2110.01799/) (2021)
- [CLAUDETTE: an Automated Detector of Potentially Unfair Clauses in Online Terms of Service](https://mlsanity.com/paper/1805.01217/) (2018)
- [Question Answering for Privacy Policies: Combining Computational and Legal Perspectives](https://mlsanity.com/paper/1911.00841/) (2019)
- [Polisis: Automated Analysis and Presentation of Privacy Policies Using Deep Learning](https://mlsanity.com/paper/1802.02561/) (2018)
- [Privacy Policies over Time: Curation and Analysis of a Million-Document Dataset](https://mlsanity.com/paper/2008.09159/) (2020)
- [Dark Patterns after the GDPR: Scraping Consent Pop-ups and Demonstrating their Influence](https://mlsanity.com/paper/2001.02479/) (2020)
- [Privacy at Scale: Introducing the PrivaSeer Corpus of Web Privacy Policies](https://mlsanity.com/paper/2004.11131/) (2020)
- [The Privacy Policy Landscape After the GDPR](https://mlsanity.com/paper/1809.08396/) (2018)
- [Large Language Models: A New Approach for Privacy Policy Analysis at Scale](https://mlsanity.com/paper/2405.20900/) (2024)
- [ContraDoc: Understanding Self-Contradictions in Documents with Large Language Models](https://mlsanity.com/paper/2311.09182/) (2023)
- [Automatic Detection of Vague Words and Sentences in Privacy Policies](https://mlsanity.com/paper/1808.06219/) (2018)
- [PoliGraph: Automated Privacy Policy Analysis using Knowledge Graphs (Journal Version)](https://mlsanity.com/paper/2210.06746/) (2022)
- [AI-enabled Automation for Completeness Checking of Privacy Policies](https://mlsanity.com/paper/2106.05688/) (2021)
- [ATLAS: Automatically Detecting Discrepancies Between Privacy Policies and Privacy Labels](https://mlsanity.com/paper/2306.09247/) (2023)
- [Explainable Compliance Detection with Multi-Hop Natural Language Inference on Assurance Case Structure](https://mlsanity.com/paper/2506.08713/) (2025)
- [PolicyLR: A Logic Representation For Privacy Policies](https://mlsanity.com/paper/2408.14830/) (2024)
- [Detecting Greenwashing: A Natural Language Processing Literature Survey](https://mlsanity.com/paper/2502.07541/) (2025)
- [Anonymity-washing](https://mlsanity.com/paper/2505.18627/) (2025)
- [Jurisdiction as Structural Barrier: How Privacy Policy Organization May Reduce Visibility of Substantive Disclosures](https://mlsanity.com/paper/2601.20792/) (2026)

## Abstract (arXiv)

Privacy policies may contain internal contradictions in which commitments are undermined by practices documented elsewhere in the same policy. We operationalize this phenomenon, privacy washing, through a four-stage pipeline: statement extraction, compatibility filtering and natural language inference screening, multi-model judge verification, and thematic analysis, with contradictions confirmed by majority vote of a three-model LLM panel. Applied to two corpora of website privacy policies, 123 collected in 2026 (OPPT) and 115 collected in 2015 (OPP-115), the pipeline finds the same category patterns recurring across the 11-year gap, with third-party sharing contradictions the majority of confirmed cases in each primary run, consistent with structural factors in policy composition rather than necessarily intentional deception. At least one panel-confirmed contradiction appears in 12.2% of OPPT companies (15/123; 9.8% excluding legacy pairs) and 36.5% of OPP-115 companies (42/115). A stability re-run seven months later, with a fully separated configuration (new extraction models, judges from three Chinese providers absent from both corpora, matched filters, no judge-submission similarity threshold), reproduces the OPPT prevalence under the original protocol (13.0% vs. 12.2%), finds sub-threshold pairs confirm at rates of the same order as those above (raising prevalence to 20.3% and 40.9%), and shows the third-party majority is panel-sensitive while the recurrence of the same category pairs is not. Two caveats govern all figures: panel verdicts are not validated against human expert judgment, so precision is unknown and prevalence figures are lower bounds; and the two primary runs used different filter configurations, so their prevalence difference is not interpretable as a corpus or era effect (the matched re-run reduces the gap to roughly twofold but does not eliminate it).