AI-Assisted Rewrite of Open Source Code Sparks Licensing Debate

A recent update to the open source Python library chardet has sparked debate over the legal and ethical implications of using artificial intelligence to rewrite software and potentially alter its licensing terms. The episode highlights increasing questions facing the tech industry as developers turn to AI to refactor existing open source projects.

Chardet, originally written by Mark Pilgrim in 2006, is widely used to automatically detect the character encoding of text files. Pilgrim released the initial code under the GNU Lesser General Public License (LGPL), a license that restricts how code may be reused and redistributed. The goal of such licenses is to ensure software remains open and subject to certain user freedoms.

In 2012, Dan Blanchard assumed maintenance of the chardet repository. Last week, Blanchard announced the release of version 7.0, describing it as a "ground-up, MIT-licensed rewrite" of the library. This new iteration was built with the help of Claude Code, an artificial intelligence coding assistant developed by Anthropic. The update aimed to deliver improved performance and accuracy, but it also introduced a significant change: licensing the code under the less restrictive MIT license.

The central question this development raises is whether using AI to generate new code, in essence a clean-room rewrite, is sufficient to sidestep prior licensing restrictions. Traditionally, "reverse engineering" and clean-room approaches have been accepted methods for replicating functionality without directly copying protected code. However, the use of large language models (LLMs)—AI systems trained on vast quantities of text—to help rewrite code adds a new layer of complexity to legal and ethical considerations.

Some legal and open source experts have expressed concerns that using AI tools for such rewrites might not fully eliminate dependencies on the original code, raising uncertainty over compliance with original licenses. They caution that without careful safeguards and clear documentation of the rewrite process, AI-assisted code could inadvertently replicate protected elements, blurring the line between original and derived works.

The controversy surrounding chardet is part of a broader conversation about the impact of generative AI on open source software development. As AI tools become more powerful and accessible, their use in software engineering is likely to challenge established legal frameworks and norms around software licensing, copyright, and attribution.

While the chardet case is situated in the global context, it resonates in Europe as well, where open source adoption and software licensing practices are integral to public sector infrastructure and private enterprise. The coming years are likely to see increased scrutiny from both regulators and the open source community over how AI tools are deployed in codebase rewrites and license transitions.

For developers and organizations, the episode underscores the importance of understanding both the capabilities and the limitations of AI in software engineering—especially when navigating the complex world of open source licenses.

Reference: arstechnica.com