Does AI-Assisted Development Homogenize Software? An Empirical Study of Code Similarity, Quality, and Evolution

Key points

• AI-assisted software development is evolving from interactive LLM assistance toward agentic systems that autonomously perform software engineering tasks, yet the effects of this shift remain insufficiently understood.1
• Using snapshots from Software Heritage, we examine code similarity, software quality, and software evolution.1
• Results show no consistent evidence of increased similarity as semantic and syntactic similarity decreased significantly in Java for both the LLM and agentic eras relative to the pre-LLM era, while Python showed no significant differences.1
• These findings challenge the assumption that LLM- and agent-assisted development necessarily homogenizes code, degrades quality, or accelerates growth.1

Sources (1)

• [1]Does AI-Assisted Development Homogenize Software? An Empirical Study of Code Similarity, Quality, and Evolution
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 16, 06:22 PM

  We present an empirical study of 50 long-lived Java and Python open-source projects across three eras: pre-LLM, potentially LLM-assisted, and detected agentic development.

  AI-assisted software development is evolving from interactive LLM assistance toward agentic systems that autonomously perform software engineering tasks, yet the effects of this shift remain insufficiently understood.

Extractive summary: sentences quoted from the sources.