# The Perfection Paradox: From Architect to Curator in AI-Assisted API Design

arXiv: https://arxiv.org/abs/2603.12475 · HTML: https://mlsanity.com/paper/2603.12475/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 52 references that are in the map.
- First posted on arXiv 12 Mar 2026.

## Builds on (most cited first, 14 of 14)

- [ChainForge: A Visual Toolkit for Prompt Engineering and LLM Hypothesis Testing](https://mlsanity.com/paper/2309.09128/) (2023)
- [A Framework for Human Evaluation of Large Language Models in Healthcare Derived from Literature Review](https://mlsanity.com/paper/2405.02559/) (2024)
- [Design Principles for Generative AI Applications](https://mlsanity.com/paper/2401.14484/) (2024)
- [A Survey on Evaluating Large Language Models in Code Generation Tasks](https://mlsanity.com/paper/2408.16498/) (2024)
- [The Effects of Generative AI on Design Fixation and Divergent Thinking](https://mlsanity.com/paper/2403.11164/) (2024)
- [L2CEval: Evaluating Language-to-Code Generation Capabilities of Large Language Models](https://mlsanity.com/paper/2309.17446/) (2023)
- [Multi-line AI-assisted Code Authoring](https://mlsanity.com/paper/2402.04141/) (2024)
- [Artificial-Intelligence Generated Code Considered Harmful: A Road Map for Secure and High-Quality Code Generation](https://mlsanity.com/paper/2409.19182/) (2024)
- [Comparing Software Developers with ChatGPT: An Empirical Investigation](https://mlsanity.com/paper/2305.11837/) (2023)
- [Human and Machine: How Software Engineers Perceive and Engage with AI-Assisted Code Reviews Compared to Their Peers](https://mlsanity.com/paper/2501.02092/) (2025)
- [APITestGenie: Automated API Test Generation through Generative AI](https://mlsanity.com/paper/2409.03838/) (2024)
- [Opportunities for Large Language Models and Discourse in Engineering Design](https://mlsanity.com/paper/2306.09169/) (2023)
- [A methodology to Evaluate the Usability of Security APIs](https://mlsanity.com/paper/1810.05100/) (2018)
- [A Generic Cognitive Dimensions Questionnaire to Evaluate the Usability of Security APIs](https://mlsanity.com/paper/1703.09846/) (2017)

## Abstract (arXiv)

Enterprise API design is often bottlenecked by the tension between rapid feature delivery and the rigorous maintenance of usability standards. We present an industrial case study evaluating an AI-assisted design workflow trained on API Improvement Proposals (AIPs). Through a controlled study with 16 industry experts, we compared AI-generated API specifications against human-authored ones. While quantitative results indicated AI superiority in 10 of 11 usability dimensions and an 87% reduction in authoring time, qualitative analysis revealed a paradox: experts frequently misidentified AI work as human (19% accuracy) yet described the designs as unsettlingly "perfect." We characterize this as a "Perfection Paradox" -- where hyper-consistency signals a lack of pragmatic human judgment. We discuss the implications of this perfection paradox, proposing a shift in the human designer's role from the "drafter" of specifications to the "curator" of AI-generated patterns.