arXiv subjects

Alina Wernick

Publications and source records attributed to Alina Wernick.

4 recordsLinked to original sources

arXiv2026

Law of Large Numbers: Accuracy as Statistical Measure for AI Compliance and Competition

The machine learning community progresses (in part) by improving the "accuracy" of its systems. The EU AI Act explicitly refers to "accuracy" as part of its compliance measures for high-risk AI systems. Are we talking about the same thing? This work presents "accuracy" as a case-study for differing requirements of social worlds, the technological machine learning community and the legal community. While competition on accuracy contributes to technological development, machine learning scholars simultaneously recognize accuracy's shortcomings regarding the usefulness and effectiveness of machine learning systems. The legal counterpart embraces the vagueness of "accuracy," leaving interpretative flexibility for technological and societal changes. At the same time, accuracy is a core element of compliance within the EU AI Act. We elaborate on five main tensions, (a) nature of accuracy, (b) notion of performance, (c) scope of validity, (d) ends, and (e) statisticalness, to show that the two communities project disparate, and sometimes contradictory, expectations on accuracy. Both legal and technical communities lack precise understanding of "accuracy" beyond the contextual boundaries of their community. The resulting frictions, \eg, based on the empirical or normative understanding of accuracy, are symptoms of an unresolved (and unresolvable) debate on what accuracy is. We constructively use the frictions to recommend baselines and interventional studies in standardization, and demand for tools to extend the validity of accuracy measurements.

cs.CY↗

arXiv2025

Position: EU AI Act's Research Exemptions Can Break the Publication Norms of Major AI Conferences

The EU has become one of the vanguards in regulating the digital age. A particularly important regulation in the Artificial Intelligence (AI) domain is the 2024 enacted EU AI Act. The AI Act specifies -- due to a risk-based approach -- various obligations for providers of AI systems. These obligations, for example, include a cascade of documentation and compliance measures, which represent a potential obstacle to science. But do these obligations also apply to AI researchers? This position paper argues that, indeed, the AI Act's obligations could apply in many more cases than the AI community is aware of. Moreover, we argue that the AI Act is drafted in a manner that may unwillingly disrupt the scientific publication practices of the AI research community, with a focus on model and system release. We contribute the following: 1. We offer a high-level roadmap for AI researchers to evaluate whether they need to comply with the AI Act 2. We explain with everyday research examples why the AI Act applies to AI research. 3. We analyse the exceptions of the AI Act's applicability AI research and offer visual tool for researchers to navigate the AI Act's complex system or research exceptions 4. We establish a position the AI Act's research exceptions fail to account for current AI research conventions, as publishing AI research may void the research exceptions of the Act. 5. We propose changes to the AI Act to provide more legal certainty for AI researchers and give two recommendations for AI researchers to reduce the risk of not complying with the AI Act. We see our paper as a starting point for a discussion between policymakers, legal scholars, and AI researchers to avoid unintended side effects of the AI Act.

cs.CY↗

arXiv2025

Evaluating the Contextual Integrity of False Positives in Algorithmic Travel Surveillance

International air travel is highly surveilled. While surveillance is deemed necessary for law enforcement to prevent and detect terrorism and other serious crimes, even the most accurate algorithmic mass surveillance systems produce high numbers of false positives. Despite the potential impact of false positives on the fundamental rights of millions of passengers, algorithmic travel surveillance is lawful in the EU. However, as the system's processing practices and accuracy are kept secret by law, it is unknown to what degree passengers are accepting of the system's interference with their rights to privacy and data protection. We conducted a nationally representative survey of the adult population of Finland (N=1550) to assess their attitudes towards algorithmic mass surveillance in air travel and its potential expansion to other travel contexts. Furthermore, we developed a novel approach for estimating the threshold, beyond which, the number of false positives breaches individuals' perception of contextual integrity. Surprisingly, when faced with a trade-off between privacy and security, even very high false positive counts were perceived as legitimate. This result could be attributed to Finland's high-trust cultural context, but also raises questions about people's capacity to account for privacy harms that happen to other people. We conclude by discussing how legal and ethical approaches to legitimising algorithmic surveillance based on individual rights may overlook the statistical or systemic properties of mass surveillance.

cs.CY↗

arXiv2024

Ordre public exceptions for algorithmic surveillance patents

This chapter explores the role of patent protection in algorithmic surveillance and whether ordre public exceptions from patentability should apply to such patents, due to their potential to enable human rights violations. It concludes that in most cases, it is undesirable to exclude algorithmic surveillance patents from patentability, as the patent system is ill-equipped to evaluate the impacts of the exploitation of such technologies. Furthermore, the disclosure of such patents has positive externalities from the societal perspective by opening the black box of surveillance for public scrutiny.

cs.CY↗