RECORD
AI data transparency: an exploration through the lens of AI incidents
Computers and Society · Submitted 5 Sep 2024
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Our findings demonstrate that low data transparency persists across a wide range of systems, and further that issues of transparency and explainability at model- and system- level create barriers for investigating data transparency information to address public concerns about AI systems.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Knowing more about the data used to build AI systems is critical for allowing different stakeholders to play their part in ensuring responsible and appropriate deployment and use. Meanwhile, a 2023 report shows that data transparency lags significantly behind other areas of AI transparency in popular foundation models. In this research, we sought to build on these findings, exploring the status of public documentation about data practices within AI systems generating public concern. Our findings demonstrate that low data transparency persists across a wide range of systems, and further that issues of transparency and explainability at model- and system- level create barriers for investigating data transparency information to address public concerns about AI systems. We highlight a need to develop systematic ways of monitoring AI data transparency that account for the diversity of AI system types, and for such efforts to build on further understanding of the needs of those both supplying and using data transparency information.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Knowing more about the data used to build AI systems is critical for allowing different stakeholders to play their part in ensuring responsible and appropriate deployment and use. | background 1.00 | 0.04 | 0.04 | asserted 0.99 |
| s1 | Meanwhile, a 2023 report shows that data transparency lags significantly behind other areas of AI transparency in popular foundation models. | background 0.94 | 0.04 | 0.03 | asserted 1.00 |
| s2 | In this research, we sought to build on these findings, exploring the status of public documentation about data practices within AI systems generating public concern. | contribution 0.99 | 0.68 | 0.03 | asserted 0.69 |
| s3 | Our findings demonstrate that low data transparency persists across a wide range of systems, and further that issues of transparency and explainability at model- and system- level create barriers for investigating data transparency information to address public concerns about AI systems. | result 0.98 | 0.95 | 0.82 | asserted 1.00 |
| s4 | We highlight a need to develop systematic ways of monitoring AI data transparency that account for the diversity of AI system types, and for such efforts to build on further understanding of the needs of those both supplying and using data transparency information. | future work 1.00 | 0.11 | 0.32 | asserted 0.76 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Knowing more about the data used to build AI systems is critical for allowing different stakeholders to play their part in ensuring responsible and appropriate deployment and use. Meanwhile, a 2023 report shows that data transparency lags significantly behind other areas of AI transparency in popular foundation models. In this research, we sought to build on these findings, exploring the status of public documentation about data practices within AI systems generating public concern. Our findings demonstrate that low data transparency persists across a wide range of systems, and further that issues of transparency and explainability at model- and system- level create barriers for investigating data transparency information to address public concerns about AI systems. We highlight a need to develop systematic ways of monitoring AI data transparency that account for the diversity of AI system types, and for such efforts to build on further understanding of the needs of those both supplying and using data transparency information.
Human review
How the literature uses it
1 arXiv paper cite this in 1 sentence
Papers citing it
Cites it as support 1 paper
-
Additionally, even if users were to articulate harms as they occur, the lack of transparency in current AI systems [ref] and the difficulty for users to steer AI system behavior [ref] make it unclear whether this labor would prevent harms in future outputs.
Identifying Harm in Personalized, Generative AI Systems Requires User-Centered Auditing at the Interaction Level · Asymmetry of Labor · 0.99
What it cites · 3 sentences without a confident label
Uses the method or tool of 1 paper
-
In assessing ‘data transparency’, we used the list of indicators under the ‘Data’ section (one of ten transparency areas in their ‘ecosystem approach’ to AI transparency, see figure 1) in the Stanford Foundation Model Transparency Index (FMTI), developed from their review of scientific literature and public discourse on what AI transparency is needed for “public accountability” (p. 28, [ref] ) among foundation models.
The Foundation Model Transparency Index · Analysing AI data transparency · 0.95
Cites as support 5 papers
-
This is corroborated by recent research into transparency of the most popular AI datasets [ref] and the Stanford research exploring AI lifecycle transparency within high profile foundation models ( [ref] , see figure 1), which have demonstrated low availability of data transparency information in practice.
The Foundation Model Transparency Index · Where are we now? · 1.00
-
This is corroborated by recent research into transparency of the most popular AI datasets [ref] and the Stanford research exploring AI lifecycle transparency within high profile foundation models ( [ref] , see figure 1), which have demonstrated low availability of data transparency information in practice.
The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI · Where are we now? · 1.00
-
However, while recent studies have suggested that both data- and model- documentation practices are increasingly used, particularly among developer communities, evidence suggests that information supplied within documents is often limited and highly inconsistent [ref] .
Navigating Dataset Documentations in AI: A Large-Scale Analysis of Dataset Cards on Hugging Face · Where are we now? · 0.99
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.