#algorithmic-accountability News & Analysis

4 articles tagged with #algorithmic-accountability. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

4 articles

AIBearisharXiv – CS AI · 2d ago7/10

🧠

The Illusion of Opting in AI-Mediated Consequential Decisions

A new academic framework argues that AI systems create an 'illusion of opting'—where users appear to have meaningful choice while their actual decision-making agency is systematically weakened. The research proposes three normative imperatives (existential honesty, ecological rationality, and counterfactual reparation) to protect human agency in AI-mediated consequential decisions, particularly for vulnerable populations.

AINeutralarXiv – CS AI · Apr 147/10

🧠

Regional Explanations: Bridging Local and Global Variable Importance

Researchers identify fundamental flaws in Local Shapley Values and LIME, two widely-used machine learning interpretation methods that fail to reliably detect locally important features. They propose R-LOCO, a new approach that bridges local and global explanations by segmenting input space into regions and applying global attribution methods within those regions for more faithful local attributions.

AIBearisharXiv – CS AI · Apr 206/10

🧠

Bureaucratic Silences: What the Canadian AI Register Reveals, Omits, and Obscures

Canada's new Federal AI Register, designed to enhance transparency, reveals that 86% of deployed AI systems serve internal efficiency purposes while systematically obscuring crucial details about human oversight, training data, and decision-making uncertainty. Researchers analyzing the 409-system dataset found the register prioritizes technical descriptions over sociotechnical context, potentially transforming accountability into performative compliance rather than genuine contestability.

AINeutralarXiv – CS AI · Apr 146/10

🧠

AI Integrity: A New Paradigm for Verifiable AI Governance

Researchers introduce AI Integrity, a new governance framework that verifies the reasoning processes of AI systems rather than just evaluating outcomes. The approach defines an Authority Stack—a four-layer model of values, epistemological standards, source preferences, and data criteria—and proposes the PRISM framework to measure integrity through six core metrics, addressing a critical gap in existing AI Ethics, Safety, and Alignment paradigms.