Generative AI for Managerial Decision-Making under Ambiguity and Sycophancy
Not on the map. Why?
Abstract (arXiv)
Generative artificial intelligence (GenAI) is increasingly being integrated into complex business workflows, fundamentally shifting the boundaries of managerial decision-making. However, the reliability of its strategic advice in ambiguous business contexts remains a critical knowledge gap. To address this gap, this study compares multiple GenAI models in their ability to detect ambiguity, examines whether a systematic ambiguity-resolution process improves response quality, and investigates their susceptibility to sycophantic behavior when confronted with flawed managerial directives. Using a novel four-dimensional business ambiguity taxonomy, we conducted a human-in-the-loop experiment across strategic, tactical, and operational scenarios. The resulting decisions were assessed through a human-validated automated evaluation framework based on agreement, actionability, justification quality, and constraint adherence. The results show that our approach not only distinguishes different types of ambiguity, but also reveals how ambiguity resolution systematically changes model behavior. In particular, resolving ambiguities improved decision quality across all managerial levels, with the strongest gains observed in constraint adherence. The analysis further showed that sycophantic behavior is not uniform across models: some models challenged flawed assumptions, whereas others tended to comply with them. This study contributes to the bounded rationality literature by positioning GenAI as a cognitive scaffold that can detect and resolve ambiguities managers might overlook, while demonstrating that its artificial limitations require human oversight to ensure its reliability as a strategic partner.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 39 of its 137 references that are in the map. First posted on arXiv 4 Mar 2026.
Builds on
39 of 137 references.
References · 39
- 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
- 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
- 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,592
- 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
- 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
- 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingcited by 1,122
- 2023Large Language Models are not Fair Evaluatorscited by 1,085
- 2023Towards Understanding Sycophancy in Language Modelscited by 1,040
- 2024Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluatorscited by 897
- 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 836
- 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 698
- 2024LLM Evaluators Recognize and Favor Their Own Generationscited by 686
27 more
- 2023A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPTcited by 619
- 2024LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methodscited by 465
- 2020AmbigQA: Answering Ambiguous Open-domain Questionscited by 390
- 2023Simple synthetic data reduces sycophancy in large language modelscited by 187
- 2025SycEval: Evaluating LLM Sycophancycited by 108
- 2024Sycophancy in Large Language Models: Causes and Mitigationscited by 101
- 2023When Large Language Models contradict humans? Large Language Models' Sycophantic Behaviourcited by 62
- 2024Aligning Language Models to Explicitly Handle Ambiguitycited by 56
- 2021Hybrid Intelligencecited by 55
- 2024CLAMBER: A Benchmark of Identifying and Clarifying Ambiguous Information Needs in Large Language Modelscited by 54
- 2024Beyond correlation: The Impact of Human Uncertainty in Measuring the Effectiveness of Automatic Evaluation and LLM-as-a-Judgecited by 31
- 2024Linear Probe Penalties Reduce LLM Sycophancycited by 28
- 2025Collaborating with AI Agents: Field Experiments on Teamwork, Productivity, and Performancecited by 14
- 2023AmbiCoref: Evaluating Human and Model Sensitivity to Ambiguous Coreferencecited by 12
- 2023Self-Consistency of Large Language Models under Ambiguitycited by 12
- 2023Ambiguity-Aware In-Context Learning with Large Language Modelscited by 11
- 2024Scope Ambiguities in Large Language Modelscited by 11
- 2025Towards Detecting Prompt Knowledge Gaps for Improved LLM-guided Issue Resolutioncited by 11
- 2025The Confidence Paradox: Can LLM Know When It's Wrongcited by 11
- 2025AmbiSQL: Interactive Ambiguity Detection and Resolution for Text-to-SQLcited by 11
- 2025Clarifying Ambiguities: on the Role of Ambiguity Types in Prompting Methods for Clarification Generationcited by 10
- 2024Artificial Intelligence and Strategic Decision-Making: Evidence from Entrepreneurs and Investorscited by 9
- 2021Annotation Curricula to Implicitly Train Non-Expert Annotatorscited by 8
- 2024A Taxonomy of Ambiguity Types for NLPcited by 7
- 2025Iterative Resolution of Prompt Ambiguities Using a Progressive Cutting-Search Approachcited by 2
- 2015Strategic and Operational information support of decision making processes and systemscited by 1
- 2025Ambiguity Resolution with Human Feedback for Code Writing Taskscited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.