Explanation in AI and law: Past, present and future
Open accessThe DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index.
Abstract
Abstract Explanation has been a central feature of AI systems for legal reasoning since their inception. Recently, the topic of explanation of decisions has taken on a new urgency, throughout AI in general, with the increasing deployment of AI tools and the need for lay users to be able to place trust in the decisions that the support tools are recommending. This paper provides a comprehensive review of the variety of techniques for explanation that have been developed in AI and Law. We summarise the early contributions and how these have since developed. We describe a number of notable current methods for automated explanation of legal reasoning and we also highlight gaps that must be addressed by future systems to ensure that accurate, trustworthy, unbiased decision support can be provided to legal professionals. We believe that insights from AI and Law, where explanation has long been a concern, may provide useful pointers for future development of explainable AI.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 1×
“Specifically, HANs use bi-directional gated recurrent units (GRUs) [49] over pre-trained static word embeddings and concatenate the forward and backward hidden states for each token as a contextualised word representation.”
BackgroundExplaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
Xu, K., Ba, J., Kiros, R., Cho, K., Courville, A., Salakhutdinov, R., Zemel, R., & Bengio, Y. (2015) · arXiv · doi:10.48550/arxiv.1502.03044 · Cited by 8,248
Toulmin, S. E. (2003) · doi:10.1017/cbo9780511840005 · Cited by 7,101
Explanation in Artificial Intelligence: Insights from the Social Sciences
Miller, T. (2017) · arXiv · doi:10.48550/arxiv.1706.07269 · Cited by 5,338
Explanation in artificial intelligence: Insights from the social sciences
Miller, T. (2019) · Artificial Intelligence · doi:10.1016/j.artint.2018.07.007 · Cited by 5,302
Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1406.1078 · Cited by 3,317
Self-Training With Noisy Student Improves ImageNet Classification
Xie, Q., Luong, M.-T., Hovy, E., & Le, Q. V. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.01070 · Cited by 2,393
Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
Bolukbasi, T., Chang, K.-W., Zou, J., Saligrama, V., & Kalai, A. (2016) · arXiv · doi:10.48550/arxiv.1607.06520 · Cited by 1,615 · cited 1×
“For example, automatically learnt word representations using deep learning approaches such as word2vec [89] and Global Vector Prediction (GloVe) [94] have shown to encode un-fair discriminative gender and racial biases [142, 39].”
MethodFeature selection, L 1 vs. L 2 regularization, and rotational invariance
Ng, A. Y. (2004) · Twenty-first international conference on Machine learning - ICML '04 · doi:10.1145/1015330.1015435 · Cited by 1,572 · cited 1×
“The weight associated with a feature can be used as a proxy for determining the importance of that feature for making predictions [93].”
BackgroundAttention is not not Explanation
Wiegreffe, S., & Pinter, Y. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1002 · Cited by 966 · cited 1×
“However, it has been shown that attention alone is inadequate as a form of explanation especially when the number of layers in a deep neural network increases and multiple nonlinear activation functions are used after an attention layer [116, 134].”
BackgroundThe Risk of Racial Bias in Hate Speech Detection
Sap, M., Card, D., Gabriel, S., Choi, Y., & Smith, N. A. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1163 · Cited by 817 · cited 1×
“However, doing so makes those NLP systems biased with regard to legally protected attributes such as gender [112].”
BackgroundMatching the Blanks: Distributional Similarity for Relation Learning
Baldini Soares, L., FitzGerald, N., Ling, J., & Kwiatkowski, T. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1279 · Cited by 731 · cited 1×
“Moreover, state-of-the-art performance on a broad range of classification/recognition tasks have been achieved such as in image classification [139], machine translation [56], textual entailment [80] and relation extraction [13] using deep learning methods.”
BackgroundSerrano, S., & Smith, N. A. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1282 · Cited by 690 · cited 1×
“However, it has been shown that attention alone is inadequate as a form of explanation especially when the number of layers in a deep neural network increases and multiple nonlinear activation functions are used after an attention layer [116, 134].”
BackgroundExplainable Prediction of Medical Codes from Clinical Text
Mullenbach, J., Wiegreffe, S., Duke, J., Sun, J., & Eisenstein, J. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1100 · Cited by 597
The British Nationality Act as a logic program
Sergot, M. J., Sadri, F., Kowalski, R. A., Kriwaczek, F., Hammond, P., & Cory, H. T. (1986) · Communications of the ACM · doi:10.1145/5689.5920 · Cited by 564 · cited 3×
“For our example of this style of system we will consider logic programming in the style of [115], or, as applied to case law, [33].”
MethodAccountability of AI Under the Law: The Role of Explanation
Doshi-Velez, F., Kortz, M., Budish, R., Bavitz, C., Gershman, S. J., O'Brien, D., Scott, K., Shieber, S., Waldo, J., Weinberger, D., et al. (2017) · doi:10.2139/ssrn.3064761 · Cited by 457 · cited 2×
“This is the right to explanation [54].”
BackgroundGonen, H., & Goldberg, Y. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1061 · Cited by 454 · cited 1×
“Although there have been recent attempts to debias pre-trained word embeddings [75], it has been reported that not all biases are accurately removed by the existing methods [58].”
BackgroundMao, J., Gan, C., Kohli, P., Tenenbaum, J. B., & Wu, J. (2019) · arXiv · doi:10.48550/arxiv.1904.12584 · Cited by 444
Semi-Supervised Methods for Explainable Legal Prediction
Branting, K., Weiss, B., Brown, B., Pfeifer, C., Chakraborty, A., Ferro, L., Pfaff, M., & Yeh, A. (2019) · Proceedings of the Seventeenth International Conference on Artificial Intelligence and Law · doi:10.1145/3322640.3326723 · Cited by 43 · cited 4×
“However, it has been shown that highlighting salient words in legal documents alone does not significantly help to reduce the time required to make legal decisions [43].”
BackgroundMethodInfluentialNoise induced hearing loss: Building an application using the ANGELIC methodology
Al-Abdulkarim, L., Atkinson, K., Bench-Capon, T., Whittle, S., Williams, R., & Wolfenden, C. (2018) · Argument & Computation · doi:10.3233/aac-181005 · Cited by 21 · cited 4×
“To move the work described above into real world applications, a feasibility study was conducted in collaboration with a large law firm to build a practical system us-ing ANGELIC [4].”
BackgroundMethodInfluentialSelf-citation
Cited by
Nazar, M., Alam, M. M., Yafi, E., & Su'ud, M. M. (2021) · IEEE Access · doi:10.1109/access.2021.3127881 · Cited by 312
Unveiling the shadows: beyond the hype of AI in education
Al-Zahrani, A. M. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e30696 · Cited by 225
Explainable Artificial Intelligence for Tabular Data: A Survey
Sahakyan, M., Aung, Z., & Rahwan, T. (2021) · IEEE Access · doi:10.1109/access.2021.3116481 · Cited by 135
“For a more thorough discussion of XAI applications in the area of law and criminal justice, see the surveys by Deeks [190] and Atkinson et al. [189].”
BackgroundXAIR: A Systematic Metareview of Explainable AI (XAI) Aligned to the Software Development Process
Clement, T., Kemmerzell, N., Abdelaal, M., & Amberg, M. (2023) · Machine Learning and Knowledge Extraction · doi:10.3390/make5010006 · Cited by 131
Parycek, P., Schmid, V., & Novak, A.-S. (2023) · Journal of the Knowledge Economy · doi:10.1007/s13132-023-01433-3 · Cited by 131
“These systems can also be used in the field of legistics to test and simulate the enforcement of laws (Atkinson et al., 2020).”
Metadata
- Authors
- Atkinson, K., Bench-Capon, T., & Bollegala, D.
- Venue
- Artificial Intelligence
- Publisher
- Elsevier BV
- Published
- 2020-12
- Volume
- 289
- Pages
- 103387
- ISSN
- 0004-3702 (ISSN-L)
- OpenAlex
- W3086560451
- Cited by
- 178
- References
- 21
- FWCI
- 9.09
- Type
- journal article
- Language
- en
- Open access
- bronze
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.