mlsanityData to 13 Oct 2026

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

arXiv 2603.12475 ↗First posted 12 Mar 2026cs.SE cs.AI cs.HCcited by 0

Not on the map. Why?

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.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 14 of its 52 references that are in the map. First posted on arXiv 12 Mar 2026.

Builds on

14 of 52 references.

References · 14

  1. 2023ChainForge: A Visual Toolkit for Prompt Engineering and LLM Hypothesis Testingcited by 88
  2. 2024A Framework for Human Evaluation of Large Language Models in Healthcare Derived from Literature Reviewcited by 66
  3. 2024Design Principles for Generative AI Applicationscited by 57
  4. 2024A Survey on Evaluating Large Language Models in Code Generation Taskscited by 43
  5. 2024The Effects of Generative AI on Design Fixation and Divergent Thinkingcited by 37
  6. 2023L2CEval: Evaluating Language-to-Code Generation Capabilities of Large Language Modelscited by 22
  7. 2024Multi-line AI-assisted Code Authoringcited by 13
  8. 2023Comparing Software Developers with ChatGPT: An Empirical Investigationcited by 11
  9. 2024Artificial-Intelligence Generated Code Considered Harmful: A Road Map for Secure and High-Quality Code Generationcited by 11
  10. 2025Human and Machine: How Software Engineers Perceive and Engage with AI-Assisted Code Reviews Compared to Their Peerscited by 8
  11. 2024APITestGenie: Automated API Test Generation through Generative AIcited by 7
  12. 2023Opportunities for Large Language Models and Discourse in Engineering Designcited by 6
2 more
  1. 2018A methodology to Evaluate the Usability of Security APIscited by 2
  2. 2017A Generic Cognitive Dimensions Questionnaire to Evaluate the Usability of Security APIscited by 1

Built on by

Not cited yet.