{"ID":2883410,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.09297","arxiv_id":"2508.09297","title":"Biased AI improves human decision-making but reduces trust","abstract":"Current AI systems minimize risk by enforcing ideological neutrality, yet this may introduce automation bias by suppressing cognitive engagement in human decision-making. We conducted randomized trials with 2,500 participants to test whether culturally biased AI enhances human decision-making. Participants interacted with politically diverse GPT-4o variants on information evaluation tasks. Partisan AI assistants enhanced human performance, increased engagement, and reduced evaluative bias compared to non-biased counterparts, with amplified benefits when participants encountered opposing views. These gains carried a trust penalty: participants underappreciated biased AI and overcredited neutral systems. Exposing participants to two AIs whose biases flanked human perspectives closed the perception-performance gap. These findings complicate conventional wisdom about AI neutrality, suggesting that strategic integration of diverse cultural biases may foster improved and resilient human decision-making.","short_abstract":"Current AI systems minimize risk by enforcing ideological neutrality, yet this may introduce automation bias by suppressing cognitive engagement in human decision-making. We conducted randomized trials with 2,500 participants to test whether culturally biased AI enhances human decision-making. Participants interacted w...","url_abs":"https://arxiv.org/abs/2508.09297","url_pdf":"https://arxiv.org/pdf/2508.09297v3","authors":"[\"Shiyang Lai\",\"Junsol Kim\",\"Nadav Kunievsky\",\"Yujin Potter\",\"James Evans\"]","published":"2025-08-12T19:20:43Z","proceeding":"cs.HC","tasks":"[\"cs.HC\",\"cs.AI\",\"cs.CY\"]","methods":"[]","has_code":false}