"TODO: Fix the Mess Gemini Created": Towards Understanding GenAI-Induced Self-Admitted Technical Debt

2026-01-12

AI Summary

This study addresses the emergence of a novel form of self-admitted technical debt (SATD) induced by generative artificial intelligence in software development, wherein developers explicitly acknowledge flaws or uncertainties in AI-generated code through source code comments. The work proposes the concept of “Generative AI–Induced Self-admitted Technical Debt” (GIST) and establishes a new theoretical framework for understanding technical debt in AI-integrated development contexts. Through large-scale mining of 6,540 GitHub comments referencing large language models in Python and JavaScript projects, complemented by qualitative content analysis, the study identifies 81 instances of GIST. Findings reveal that developers frequently incur such debt due to deferred testing, incomplete adaptation of AI outputs, and limited comprehension of generated code, thereby confirming the prevalence of GIST and elucidating its underlying mechanisms.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

As large language models (LLMs) such as ChatGPT, Copilot, Claude, and Gemini become integrated into software development workflows, developers increasingly leave traces of AI involvement in their code comments. Among these, some comments explicitly acknowledge both the use of generative AI and the presence of technical shortcomings. Analyzing 6,540 LLM-referencing code comments from public Python and JavaScript-based GitHub repositories (November 2022-July 2025), we identified 81 that also self-admit technical debt(SATD). Developers most often describe postponed testing, incomplete adaptation, and limited understanding of AI-generated code, suggesting that AI assistance affects both when and why technical debt emerges. We term GenAI-Induced Self-admitted Technical debt (GIST) as a proposed conceptual lens to describe recurring cases where developers incorporate AI-generated code while explicitly expressing uncertainty about its behavior or correctness.

Problem

• Generative AI
• Technical Debt
• Self-Admitted Technical Debt
• Code Comments
• Software Development

Innovation

• GenAI-Induced Technical Debt
• Self-Admitted Technical Debt
• Large Language Models
• AI-assisted Development
• Code Comments