Lost in Code Generation: Reimagining the Role of Software Models in AI-driven Software Engineering

2025-11-04

AI Summary

Generative AI has lowered software development barriers, yet AI-generated systems suffer from severe deficiencies in robustness, security, and maintainability—exposing a critical gap between prototypes and production-grade software. To bridge this gap, we propose the *post-hoc model reconstruction* paradigm, which transcends the traditional role of software models as static design blueprints by positioning reverse-engineered high-level models as human-AI collaboration intermediaries and evolution governance hubs. Our approach integrates natural language processing, program analysis, and model-driven engineering to automatically reconstruct semantically consistent, executable software models from AI-generated code. Empirical evaluation demonstrates significant improvements in code comprehensibility, vulnerability traceability, and change adaptability. This work establishes the first sustainable framework for engineering AI-generated software—one that balances theoretical rigor with practical feasibility.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Humans and AI: Other Foundations of Human Computation & AI
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Generative AI enables rapid ``vibe coding,"where natural language prompts yield working software systems. While this lowers barriers to software creation, it also collapses the boundary between prototypes and engineered software, leading to fragile systems that lack robustness, security, and maintainability. We argue that this shift motivates a reimagining of software models. Rather than serving only as upfront blueprints, models can be recovered post-hoc from AI-generated code to restore comprehension, expose risks, and guide refinement. In this role, models serve as mediators between human intent, AI generation, and long-term system evolution, providing a path toward sustainable AI-driven software engineering.

Problem

• Reimagining software models for AI-generated code robustness
• Addressing fragility in AI-driven software prototypes and systems
• Using models to mediate human intent and AI generation

Innovation

• Recovering models post-hoc from AI-generated code
• Models mediate human intent and AI generation
• Models guide system evolution for sustainability