The integration of artificial intelligence into the legislative process is among the most significant—and least examined—constitutional developments of the decade. This article’s concern is not the composition of legislative text ex novo, but the function constitutional theory assigns to parliaments: the scrutiny of the texts and amendments tabled before them. The research question is accordingly how the deployment of AI in the amendment phase affects the capacity of parliaments to scrutinize legislation—and the accountability, transparency, and separation of powers it secures—and whether human oversight suffices to preserve it. The inquiry centres on two Italian systems at opposite ends of a governance spectrum. GEM (Gestore EMendamenti), the Senate’s amendment-management ecosystem, operational since 2016, ranks among the most sophisticated parliamentary AI tools in use; its functions concern the processing, clustering, ordering, and prediction of amendments, not autonomous drafting—a narrow but constitutionally consequential scope. GENAI4LEX-B, a hybrid architecture combining symbolic reasoning with generative models, has been selected for the Chamber of Deputies but not yet deployed; its value is not empirical but architectural, as a counter-model in which safeguards were designed ex ante rather than emerging reactively from practice. The article advances two arguments. First, it identifies four constitutional fault lines in AI-assisted amendment processing—consequential algorithmic error, deliberate disruption through AI-generated flooding, accountability opacity, and epistemic homogenization across branches—while recognizing the countervailing potential to strengthen scrutiny; it then tests human oversight against the empirical literature on automation bias, concluding that nominal oversight is necessary but not sufficient. Secondly, it contends that the European Union AI Act leaves a significant gap for the legislative process, and proposes a five-element governance framework to ensure that the distinction between algorithmic assistance and algorithmic authorship is maintained through enforceable standards rather than self-regulation alone.
Datafication has become a central concern in debates on artificial intelligence (AI). In these discussions, the European approach to AI is often portrayed as a ‘third way’ that balances public interests with innovation. This article examines parliamentary discourse on a police project in Hamburg, Germany, to show how the seemingly value-pluralistic ‘third way’ compromise is mobilized to justify the extraction of data to test and train an AI system for CCTV surveillance. Drawing on the pragmatist “economy of conventions,” the article introduces the original notion of
data sacrifices
. The concept allows critical data studies to examine how the diverse sociomaterial costs and reductionist generalizations of formatting the world into data are linked to the dynamics of justification and critique that shape the moral conditions of possibility for data extraction. In the article, Critical Discourse Analysis is used to reconstruct five specific suborders of worth invoked in the parliamentary debate to (de-)legitimize the AI system: security, individual freedom, social justice, automation, and experimentalism. It is demonstrated how the ‘third way’ works as a specific mode of justification in which civic values such as equality and privacy are internalized to legitimize the industrialization of surveillance. Claiming an effective technological compromise for complex moral stakes realigns key political actors and watchdogs in support of data extraction. However, Critical Discourse Analysis also demonstrates how this compromise silences certain civic critics to legitimize sacrifices for AI training. The interplay of internalization and exclusion of critique thereby modifies the moral underpinnings of AI surveillance. The framework provided in this article thus enables researchers to reveal the paradoxical nature of ‘third way’ compromises for data extraction. In the case studied, these paradoxes culminate in the first German law explicitly permitting the police to transfer anonymized and non-anonymized surveillance data to external partners for machine learning purposes.
Generative AI encodes the majority's way of knowing as the default infrastructure of knowledge itself as the default infrastructure of knowledge itself, and law must learn to govern at that level of model training.
This paper explores the usage of generative AI disinformation during the period of presidential and parliamentary elections in Croatia in 2024/2025. It has employed the notion of algorithmic populism as an analytical framework with the aim of understanding new communication patterns that emerged from generative AI technologies. By combining algorithmic populism with the deepfake-cheapfake spectrum developed by Paris and Donovan (2019), this study identifies 115 cases of AI-generated disinformation content. Fifteen of these cases were found during the parliamentary, and 100 during the presidential elections. Most cases were circulated on Facebook (56) and TikTok (37), while the most used AI techniques were face swapping (43) and voice synthesis (30). Analysed disinformation campaigns mostly targeted Zoran Milanović, Andrej Plenković, and Dragan Primorac. The findings indicate a transformation of algorithmic populism during elections, extending its role from the amplification of content to production and adaptation.
Mato Brautović, Marko Roško, Ivana Grkeš Tošović et al.· Politička Misao· 0 citations
This article examines how parliamentary discourse on artificial intelligence is transformed into legal and regulatory frameworks across different political contexts. It focuses on five jurisdictions—the European Union, the United States, Brazil, Kazakhstan, and the United Kingdom—where recent AI-related legislative and parliamentary developments provide a comparative basis for analysis.
The study applies a hybrid qualitative design combining a structured comparative review of academic literature, legal acts, and policy documents with a pilot critical discourse analysis of five selected parliamentary episodes from 2023 to 2026. The analysis is based on securitization theory and the concept of digital sovereignty as a discursive project.
The study identifies distinct semantic cores in each jurisdiction: “risk–fundamental rights” in the European Union, “barriers–dominance” in the United States, “inequality–high risk” in Brazil, “national code–partnership sovereignty” in Kazakhstan, and “dependency lock-in–delayed reflection” in the United Kingdom. The findings show that H2 and H3 are supported, while H1 is not confirmed in its original formulation and requires revision.
The article contributes to AI governance studies by proposing an updated five-part typology of AI regulation. It introduces the concept of “sovereignty through partnership” as an alternative to technological autarky and one-sided regulatory borrowing, and conceptualizes “delayed reflection” as a regulatory pattern among established democracies that recognize infrastructural dependency only after it has already emerged.
Elnur Beisenbayev, Bakhytzhan Bukharbayev, Zh.M. Tolen et al.· Frontiers in Political Scien...· 1 citation
Abstract The paper provides a critical analysis of the EU AI Act (Regulation 2024/1689) within the broader context of contemporary AI developments. Starting from an historical overview on the development of advanced AI systems, it moves the focus onto the intrinsic meaning of Artificial Intelligence to highlight how, despite such fascinating wording, there cannot be a shift of responsibility onto the systems themselves—as was proposed, for example, by the European Parliament resolution of 16 February 2017 with recommendations to the Commission on Civil Law Rules on Robotics (EUR-Lex - 52017IP0051); not until, at least, singularity is achieved. Moral and legal responsibility, therefore, lies with planners, developers, implementers, and all other stakeholders who have an interest in AI systems. Through the lens of social contract theory, drawing on Hobbes’ Leviathan and recent AI ethics scholarship, the paper identifies several weaknesses within the EU AI Act itself, including the rigidity of its risk-based model, its focus on technical compliance rather than societal trust, and gaps in governance and accountability. Building on these findings, it proposes potential solutions, including dynamic risk assessment, stronger participatory mechanisms, and clearer enforcement structures. The need thus emerges to recognize and acknowledge a collective type of responsibility arising from a social contract among all parties involved. The ideal scope is to scale back individual interests in favor of the collective good. While the Act marks the first attempt to create a binding moral and legal framework for trustworthy AI among European Union Member States, its ultimate success will depend on its flexibility and on cultivating a shared sense of responsibility and partnership among all stakeholders.
Leandro Loriga· Ethics & Bioethics· 0 citations
: The growing use of artificial intelligence in China’s smart-court reform has improved judicial efficiency, case management, and consistency, but it has also raised a more fundamental question: whether AI-assisted adjudication can remain compatible with the normative foundations of judicial authority. Focusing on the Chinese context, this article examines the interaction between judicial artificial intelligence and core judicial principles, particularly judicial independence, accountability, transparency, procedural justice, neutrality, and substantive fairness. Methodologically, the study adopts normative legal analysis and qualitative interpretive inquiry based on policy documents, judicial materials, and comparative scholarship on algorithmic governance. It argues that judicial AI tools such as case similarity recommendation, judgment prediction, and deviation alerts are not merely neutral instruments of modernization. Their expanding use may reshape the boundary of judicial power, blur responsibility, weaken procedural guarantees, and reproduce bias in ways that affect adjudicative legitimacy. The article therefore contends that judicial AI should be assessed not only by efficiency gains, but also by whether it preserves the core judicial principles on which public trust depends. It further proposes stronger human control, clearer accountability structures, and more effective regulation of algorithmic opacity and bias.
Wanlu Lei, Li Li· Academic Journal of Humaniti...· 0 citations
