paper-with-me

홈 › Papers

AI, Take the Wheel: What Drives Delegation and Trust in Human-Computer Cooperative Question Answering?

2026-05-27 · Maharshi Gor, Yoo Yeon Sung, Yu Hou, Eve Fleisig, Irene Ying, Tianyi Zhou, Jordan Boyd-Graber arxiv

AI systems are fallible, and humans can make mistakes in deciding whether to trust AI over their own judgment. Thus, improving human-AI collaboration requires understanding when, why, and how humans decide to rely on AI. We study two distinct reliance decisions: the delegation choice -- deciding when to let AI act autonomously without knowing its output, and the adoption choice -- evaluating AI suggestions and deciding how to use them. Both of these decoupled reliance patterns shape collaboration, but prior work rarely studies them together in realistic settings with the same users. We address this gap by studying collaborative human--AI teams competing in a question-answering game in which humans can choose when and how to work with AI agents to win. Our 24 matches pair 23 expert humans with 16 AI agents, capturing 387 delegation and 1440 adoption decisions. While human--AI collaboration performs better than either AI or humans alone, humans make suboptimal collaboration decisions, both under-relying on correct AI suggestions (3.9% of opportunities missed) and over-relying when AI misleads them (1.7%). Both parties contribute wrong answers: reported model confidence is near chance when humans and AI disagree, while confirmation bias drives higher under-reliance (64.5%) when an AI suggestion agrees with humans' initial incorrect answer. To close this gap, we recommend calibrated confidence, evidence-grounded explanations, and mechanisms that help users refine trust.

📄 PDF Abstract BibTeX arXiv:2605.28255

Code (0)

등록된 구현이 없습니다.

Tasks

Question Answering

Similar Papers 제목 키워드 기반

Ask Not What AI Can Do, But What AI Should Do: Towards a Framework of Task Delegability

2019-02-08 · NeurIPS 2019 12 · Brian Lubars, Chenhao Tan

While artificial intelligence (AI) holds promise for addressing societal challenges, issues of exactly which tasks to automate and to what extent to do so remain understudied. We approach this problem of task delegabilit…

Incentivising Participation in Liquid Democracy with Breadth-First Delegation

2019-02-25

Liquid democracy allows members of an electorate to either directly vote over alternatives, or delegate their voting rights to someone they trust. Most of the liquid democracy literature and implementations allow each vo…

Human-AI Collaboration via Conditional Delegation: A Case Study of Content Moderation

2022-04-25 · Vivian Lai, Samuel Carton, Rajat Bhatnagar, Q. Vera Liao 외

Despite impressive performance in many benchmark datasets, AI models can still make mistakes, especially among out-of-distribution examples. It remains an open question how such imperfect models can be used effectively i…

Open-Ended Question Answering

Minimal Oversight: Uncertainty-Aware Governance for Delegated AI Systems

2026-06-04 · Carlos R. B. Azevedo arxiv

AI systems increasingly delegate decisions to specialized models, evaluators, tools, and supervisory controllers. The central AI problem is no longer only model accuracy, but uncertainty-aware governance: how much autono…

Train Tracks with Gaps: Applying the Probabilistic Method to Trains

2020-12-01 · William Kuszmaul

We identify a tradeoff curve between the number of wheels on a train car, and the amount of track that must be installed in order to ensure that the train car is supported by the track at all times. The goal is to build …