Informed Consent the Global Picture

Blain, Colin; Mackay, Margaret E.; Tanner, Judith

doi:10.1177/175045890201201101

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Reverse image search engines 7 that allow face search such as [2] have gotten remarkably and worryingly efficient in the past year. For a small fee, anyone can use their portal or their API 8 to run an automated process to uncover the "real-world" identities of the humans of ImageNet dataset. For example, in societies where sex work is socially condemned or legally criminalized, re-identification of a sex worker through image search, for example, bears a real danger for the individual victim.…”

Section: The Wordnet Effectmentioning

confidence: 99%

“…Born from World War II and the haunting and despicable practices of Nazi era experimentation [4] the 1947 Nuremberg code [84] and the subsequent 1964 Helsinki declaration [30], helped to establish the doctrine of Informed Consent which builds on the fundamental notions of human dignity and agency to control dissemination of information about oneself. This has shepherded data collection endeavors in the medical and psychological sciences concerning human subjects, including photographic data [8,56], for the past several decades. A less stringent version of informed consent, broad consent, proposed in 45 CFR 46.116(d) of the Revised Common Rule [24], has been recently introduced that still affords the basic safeguards towards protecting one's identity in large scale databases.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Large image datasets: A pyrrhic win for computer vision?

Prabhu¹,

Birhane²

2020

Preprint

View full text Add to dashboard Cite

In this paper we investigate problematic practices and consequences of large scale vision datasets. We examine broad issues such as the question of consent and justice as well as specific concerns such as the inclusion of verifiably pornographic images in datasets. Taking the ImageNet-ILSVRC-2012 dataset as an example, we perform a cross-sectional model-based quantitative census covering factors such as age, gender, NSFW content scoring, class-wise accuracy, human-cardinality-analysis, and the semanticity of the image class information in order to statistically investigate the extent and subtleties of ethical transgressions. We then use the census to help hand-curate a look-up-table of images in the ImageNet-ILSVRC-2012 dataset that fall into the categories of verifiably pornographic: shot in a non-consensual setting (up-skirt), beach voyeuristic, and exposed private parts. We survey the landscape of harm and threats both society broadly and individuals face due to uncritical and ill-considered dataset curation practices. We then propose possible courses of correction and critique the pros and cons of these. We have duly open-sourced all of the code and the census meta-datasets generated in this endeavor for the computer vision community to build on. By unveiling the severity of the threats, our hope is to motivate the constitution of mandatory Institutional Review Boards (IRB) for large scale dataset curation processes.

show abstract

Section: The Wordnet Effectmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Large image datasets: A pyrrhic win for computer vision?

Prabhu¹,

Birhane²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In Brazil, the legislation is very clear as to the right to images. According to the Federal Constitution of 1988, article 5, item X, “Intimacy, private life, honor, and the image of persons are inviolable, assuring the right to compensation for the material or moral damage resulting from its violation.” ( 8 )…”

Section: Introductionmentioning

confidence: 99%

“…5º, inciso X, “São invioláveis a intimidade, a vida privada, a honra e a imagem das pessoas, assegurado o direito a indenização pelo dano material ou moral decorrente de sua violação”. ( 8 )…”

Section: Introductionunclassified

Knowledge of healthcare professionals about rights of patient’s images

Caires

Lopes

Okuno

et al. 2015

Einstein (São Paulo)

View full text Add to dashboard Cite

ObjectiveTo assess knowledge of healthcare professionals about capture and reproduction of images of patients in a hospital setting.MethodsA cross-sectional and observational study among 360 healthcare professionals (nursing staff, physical therapists, and physicians), working at a teaching hospital in the city of São Paulo (SP). A questionnaire with sociodemographic information was distributed and data were correlated to capture and reproduction of images at hospitals.ResultsOf the 360 respondents, 142 had captured images of patients in the last year, and 312 reported seeing other professionals taking photographs of patients. Of the participants who captured images, 61 said they used them for studies and presentation of clinical cases, and 168 professionals reported not knowing of any legislation in the Brazilian Penal Code regarding collection and use of images.ConclusionThere is a gap in the training of healthcare professionals regarding the use of patient´s images. It is necessary to include subjects that address this theme in the syllabus of undergraduate courses, and the healthcare organizations should regulate this issue.

show abstract

Large image datasets: A pyrrhic win for computer vision?

Birhane

Prabhu²

2021

2021 IEEE Winter Conference on Applications of Computer Vision (WACV)

157

View full text Add to dashboard Cite

Scale the model, scale the data, scale the compute' is the reigning sentiment in the world of generative AI today. While the impact of model scaling has been extensively studied, we are only beginning to scratch the surface of data scaling and its consequences. This is especially of critical importance in the context of visionlanguage datasets such as LAION. These datasets are continually growing in size and are built based on large-scale internet dumps such as the Common Crawl, which is known to have numerous drawbacks ranging from quality, legality, and content. The datasets then serve as the backbone for large generative models, contributing to the operationalization and perpetuation of harmful societal and historical biases and stereotypes. In this paper, we investigate the effect of scaling datasets on hateful content through a comparative audit of two datasets: LAION-400M and LAION-2B. Our results show that hate content increased by nearly 12% with dataset scale, measured both qualitatively and quantitatively using a metric that we term as Hate Content Rate (HCR). We also found that filtering dataset contents based on Not Safe For Work (NSFW) values calculated based on images alone does not exclude all the harmful content in alt-text. Instead, we found that trace amounts of hateful, targeted, and aggressive text remain even when carrying out conservative filtering. We end with a reflection and a discussion of the significance of our results for dataset curation and usage in the AI community. Code and the meta-data assets curated in this paper are publicly available at https://github.com/vinayprabhu/hate_scaling. Content warning: This paper contains examples of hateful text that might be disturbing, distressing, and/or offensive.

show abstract

Informed Consent the Global Picture

Cited by 3 publications

References 0 publications

Large image datasets: A pyrrhic win for computer vision?

Large image datasets: A pyrrhic win for computer vision?

Knowledge of healthcare professionals about rights of patient’s images

Large image datasets: A pyrrhic win for computer vision?

Contact Info

Product

Resources

About