Breaking into the black box of artificial intelligence
Integrity signals
Registry
Cites
Statistics
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?
Trying to understand the meaning of the weightings and calculations in ML is very different from trying to understand other types of computer programs, such as those used in most cell phones or personal computers.
“Trying to understand the meaning of the weightings and calculations in ML is very different from trying to understand other types of computer programs, such as those used in most cell phones or personal computers, in which an expert could examine the system (as a whole) to determine what it is doing and why [151, 199].”
From The ethics of using artificial intelligence in scientific research: new guidance needed for a new tool (2025) · cites this paper as [199]
An expert computer scientist or engineer will not be able to interpret the thousands of numbers used in the weightings and the billions of calculations from those numbers in an AI/ML system.
“While an expert computer scientist or engineer could troubleshoot the code, they will not be able to interpret the thousands of numbers used in the weightings and the billions of calculations from those numbers [110, 151, 199].”
From The ethics of using artificial intelligence in scientific research: new guidance needed for a new tool (2025) · cites this paper as [199]
A prominent concern of AI models based on neural network architecture is their often criticized lack of interpretability, earning them the term 'black boxes'.
“Lastly, a prominent concern of AI models based on neural network architecture is their often criticized lack of interpretability which earned them the term ‘black boxes’27,28.”
From Large language models and multimodal foundation models for precision oncology (2024) · cites this paper as [28]
Cites
Ai, T., Yang, Z., Hou, H., Zhan, C., Chen, C., Lv, W., Tao, Q., Sun, Z., & Xia, L. (2020) · Radiology · doi:10.1148/radiol.2020200642 · Cited by 5,697 · see citing works
AI for radiographic COVID-19 detection selects shortcuts over signal
DeGrave, A. J., Janizek, J. D., & Lee, S.-I. (2021) · Nature Machine Intelligence · doi:10.1038/s42256-021-00338-7 · Cited by 631 · see citing works
Gardiner, L., Rusholme‐Pilcher, R., Colmer, J., Rees, H., Crescente, J. M., Carrieri, A. P., Duncan, S. H., Pyzer‐Knapp, E. O., Krishna, R., & Hall, A. J. W. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2103070118 · Cited by 28 · see citing works
Objective criteria for explanations of machine learning models
Yeh, C., & Ravikumar, P. (2021) · Applied AI Letters · doi:10.1002/ail2.57 · Cited by 9 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Preiksaitis, C., & Rose, C. (2023) · JMIR Medical Education · doi:10.2196/48785 · Cited by 468 · see citing works
“Moreover, concerns have arisen that generative AI may produce biased content or lack representation of all populations [8,11-15, 19, 20, 22, 24, This inability to observe how a response is generated is often referred to as “the black box” problem [49].”
BackgroundResnik, D. B., & Hosseini, M. (2025) · AI and Ethics · doi:10.1007/s43681-024-00493-8 · Cited by 228 · see citing works
Should AI models be explainable to clinicians?
Abgrall, G., Holder, A. L., Chelly Dagdia, Z., Zeitouni, K., & Monnet, X. (2024) · Critical Care · doi:10.1186/s13054-024-05005-y · Cited by 175 · see citing works
Large language models and multimodal foundation models for precision oncology
Truhn, D., Eckardt, J.-N., Ferber, D., & Kather, J. N. (2024) · npj Precision Oncology · doi:10.1038/s41698-024-00573-2 · Cited by 79 · see citing works
Hu, C., Li, L., Li, Y., Wang, F., Hu, B., & Peng, Z. (2022) · Infectious Diseases and Therapy · doi:10.1007/s40121-022-00671-3 · Cited by 63 · see citing works
Open Science at the generative AI turn: An exploratory analysis of challenges and opportunities
Hosseini, M., Horbach, S. P. J. M., Holmes, K., & Ross-Hellauer, T. (2025) · Quantitative Science Studies · doi:10.1162/qss_a_00337 · Cited by 40 · see citing works
Vakharia, P., Kufeldt, A., Meyers, M., Lane, I., & Gilpin, L. H. (2024) · Lecture Notes in Computer Science · doi:10.1007/978-3-031-71170-1_23 · Cited by 5 · see citing works
“LLMs, like many deep neural network, are considered “black box” systems; their large number of parameters makes their reasoning uninter-pretable to humans [18].”
Background
- Authors
- Savage, N.
- Corresponding author
- Savage, Neil
- ORCID
- Savage, Neil
- Venue
- Nature
- Published
- 2022-03-29
- ISSN
- 0028-0836 (ISSN-L)
- 1476-4687
- OpenAlex
- W4220670584
- Cited by
- 124
- References
- 4
- Citations by year
- 2026: 9 · 2025: 30 · 2024: 35 · 2023: 29 · 2022: 21
- FWCI
- 9.54
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Nature (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence Applications
- Keywords
- explainable AI, model interpretability, model transparency, black-box models
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
