# Effects of Generative AI Errors on User Reliance Across Task Difficulty

arXiv: https://arxiv.org/abs/2604.04319 · HTML: https://mlsanity.com/paper/2604.04319/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 49 references that are in the map.
- First posted on arXiv 5 Apr 2026.

## Builds on (most cited first, 21 of 21)

- [Explainability for Large Language Models: A Survey](https://mlsanity.com/paper/2309.01029/) (2023)
- [How do Large Language Models Handle Multilingualism?](https://mlsanity.com/paper/2402.18815/) (2024)
- [To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-assisted Decision-making](https://mlsanity.com/paper/2102.09692/) (2021)
- [Explanations Can Reduce Overreliance on AI Systems During Decision-Making](https://mlsanity.com/paper/2212.06823/) (2022)
- [When combinations of humans and AI are useful: A systematic review and meta-analysis](https://mlsanity.com/paper/2405.06087/) (2024)
- [The Generative AI Paradox: "What It Can Create, It May Not Understand"](https://mlsanity.com/paper/2311.00059/) (2023)
- [The Metacognitive Demands and Opportunities of Generative AI](https://mlsanity.com/paper/2312.10893/) (2023)
- [Bridging the Gulf of Envisioning: Cognitive Design Challenges in LLM Interfaces](https://mlsanity.com/paper/2309.14459/) (2023)
- [Conceptual Metaphors Impact Perceptions of Human-AI Collaboration](https://mlsanity.com/paper/2008.02311/) (2020)
- [Cocoa: Co-Planning and Co-Execution with AI Agents](https://mlsanity.com/paper/2412.10999/) (2024)
- [Explanations, Fairness, and Appropriate Reliance in Human-AI Decision-Making](https://mlsanity.com/paper/2209.11812/) (2022)
- [Do Large Language Models Perform the Way People Expect? Measuring the Human Generalization Function](https://mlsanity.com/paper/2406.01382/) (2024)
- [A Diachronic Perspective on User Trust in AI under Uncertainty](https://mlsanity.com/paper/2310.13544/) (2023)
- [Generation Probabilities Are Not Enough: Uncertainty Highlighting in AI Code Completions](https://mlsanity.com/paper/2302.07248/) (2023)
- [Overcoming Algorithm Aversion: A Comparison between Process and Outcome Control](https://mlsanity.com/paper/2303.12896/) (2023)
- [HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants](https://mlsanity.com/paper/2509.08494/) (2025)
- [Prompt Adaptation as a Dynamic Complement in Generative AI Systems](https://mlsanity.com/paper/2407.14333/) (2024)
- [What's Producible May Not Be Reachable: Measuring the Steerability of Generative Models](https://mlsanity.com/paper/2503.17482/) (2025)
- [How Accurate Does It Feel? -- Human Perception of Different Types of Classification Mistakes](https://mlsanity.com/paper/2302.06413/) (2023)
- [A Model of Artificial Jagged Intelligence](https://mlsanity.com/paper/2601.07573/) (2026)
- [Examining the Effects of Human-Likeness of Avatars on Emotion Perception and Emotion Elicitation](https://mlsanity.com/paper/2508.01743/) (2025)

## Built on by (most cited first, 1 of 1)

- [Method, Mind, and Morality: How People Make Sense of Artificial Intelligence](https://mlsanity.com/paper/2608.24748/) (2026)

## Abstract (arXiv)

The capabilities of artificial intelligence (AI) lie along a jagged frontier, where AI systems surprisingly fail on tasks that humans find easy and succeed on tasks that humans find hard. To investigate user reactions to this phenomenon, we developed an incentive-compatible experimental methodology based on diagram generation tasks, in which we induce errors in generative AI output and test effects on user reliance. We demonstrate the interface in a preregistered 3x2 experiment (N = 577) with error rates of 10%, 30%, or 50% on easier or harder diagram generation tasks. We confirmed that observing more errors reduces use, but we unexpectedly found that easy-task errors did not significantly reduce use more than hard-task errors, suggesting that people are not averse to jaggedness in this experimental setting. We encourage future work that varies task difficulty at the same time as other features of AI errors, such as whether the jagged error patterns are easily learned.