Audits are critical mechanisms for identifying the risks and limitations of deployed artifcial intelligence (AI) systems.However, the efective execution of AI audits remains incredibly difcult, and practitioners often need to make use of various tools to support their eforts.Drawing on interviews with 35 AI audit practitioners and a landscape analysis of 435 tools, we compare the current ecosystem of AI audit tooling to practitioner needs.While many tools are designed to help set standards and evaluate AI systems, they often fall short in supporting accountability.We outline challenges practitioners faced in their eforts to use AI audit tools and highlight areas for future tool development beyond evaluationfrom harms discovery to advocacy.We conclude that the available resources do not currently support the full scope of AI audit practitioners' needs and recommend that the feld move beyond tools for just evaluation and towards more comprehensive infrastructure for AI accountability.

`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 vision-language 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.

Complex adaptive systems such as human behaviour and social systems are inherently dynamic, messy, ambiguous, incompressible, non-determinable, and non-predictable. Due to their incompressibility, neither datasets nor models can capture complex systems in their entirety. Instead, large scale datasets and predictive models pick-up societal and historical stereotypes and injustices. Yet, machine learning tools that sort, categorize, and predict the social sphere have become common place, developed and deployed in various domains from education, law enforcement, to medicine and border control. In this talk, I emphasize that machine learning models of social systems are at best a snapshot of a moving target, and at worst systems that encode the status quo and exacerbate racism, sexism, and historical injustice. I first review some of the robust body of work that has emerged in this space. I, then, argue that equitable algorithmic systems require we look beyond technical solutions and call for acknowledgement and in depth understanding of complexity, historical injustice, and current societal power asymmetries. As well as challenging simple fairness metrics when algorithmic systems fail, broader structural rethinking, deeper understanding and envisioning an alternative future -- one that is based on current realities of machine learning systems and their impacts -- are critical.

In this paper, we audit saliency cropping algorithms used by Twitter, Google and Apple to investigate issues pertaining to the male-gaze cropping phenomenon as well as race-gender biases that emerge in post-cropping survival ratios of face-images constituting 3 × 1 grid images. In doing so, we present the first formal empirical study which suggests that the worry of a male-gaze-like image cropping phenomenon on Twitter is not at all far-fetched and it does occur with worryingly high prevalence rates in real-world full-body single-female-subject images shot with logo-littered backdrops. We uncover that while all three saliency cropping frameworks considered in this paper do exhibit acute racial and gender biases, Twitter’s saliency cropping framework uniquely elicits high male-gaze cropping prevalence rates. In order to facilitate reproducing the results presented here, we are open-sourcing both the code and the datasets that we curated at shorturl.at/iuzK9. We hope the computer vision community and saliency cropping researchers will build on the results presented here and extend these investigations to similar frameworks deployed in the real world by other companies such as Microsoft and Facebook.

In this paper we investigate problematic practices and consequences of large scale vision datasets (LSVDs). 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 the society at large and individuals face due to uncritical and ill-considered dataset curation practices. We then propose possible courses of correction and critique their pros and cons. 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.

On the one hand, complexity science and enactive and embodied cognitive science approaches emphasize that people, as complex adaptive systems, are ambiguous, indeterminable, and inherently unpredictable. On the other, Machine Learning (ML) systems that claim to predict human behaviour are becoming ubiquitous in all spheres of social life. I contend that ubiquitous Artificial Intelligence (AI) and ML systems are close descendants of the Cartesian and Newtonian worldview in so far as they are tools that fundamentally sort, categorize, and classify the world, and forecast the future. Through the practice of clustering, sorting, and predicting human behaviour and action, these systems impose order, equilibrium, and stability to the active, fluid, messy, and unpredictable nature of human behaviour and the social world at large. Grounded in complexity science and enactive and embodied cognitive science approaches, this article emphasizes why people, embedded in social systems, are indeterminable and unpredictable. When ML systems "pick up" patterns and clusters, this often amounts to identifying historically and socially held norms, conventions, and stereotypes. Machine prediction of social behaviour, I argue, is not only erroneous but also presents real harm to those at the margins of society.

The 'robot rights' debate, and its related question of 'robot responsibility', invokes some of the most polarized positions in AI ethics. While some advocate for granting robots rights on a par with human beings, others, in a stark opposition argue that robots are not deserving of rights but are objects that should be our slaves. Grounded in post-Cartesian philosophical foundations, we argue not just to deny robots 'rights', but to deny that robots, as artifacts emerging out of and mediating human being, are the kinds of things that could be granted rights in the first place. Once we see robots as mediators of human being, we can understand how the 'robots rights' debate is focused on first world problems, at the expense of urgent ethical concerns, such as machine bias, machine elicited human labour exploitation, and erosion of privacy all impacting society's least privileged individuals. We conclude that, if human being is our starting point and human welfare is the primary concern, the negative impacts emerging from machinic systems, as well as the lack of taking responsibility by people designing, selling and deploying such machines, remains the most pressing ethical discussion in AI.