Tag
This paper introduces a framework using generative AI agents to automate black-box audits of personalization algorithms, demonstrating with 1,120 agents on X after the 2024 U.S. election that the algorithmic feed amplifies toxic, polarizing, and right-leaning content compared to the chronological feed.
This paper analyzes Canada's Federal AI Register (409 systems) and argues that such transparency artifacts configure accountability through ontological design rather than enabling genuine contestability, finding that 86% of systems are internal-efficiency focused while human discretion is systematically obscured.