AI-Powered Rewrite of Popular Code Library Ignites Licensing Debate

A routine software update has sparked a complex debate over copyright, artificial intelligence, and the spirit of open-source development. The controversy centers on chardet, a widely used Python library for detecting text character encoding.

Last week, maintainer Dan Blanchard released version 7.0, describing it as a complete, ground-up rebuild of the tool. The new library, he reported, is 48 times faster and more accurate. Crucially, Blanchard also changed its license from the restrictive LGPL to the more permissive MIT license, which allows for use in proprietary software. He accomplished this overhaul in about five days with significant assistance from Claude Code, an AI programming tool.

Blanchard told The Register the changes were necessary to potentially include chardet in Python's standard library, a goal long hindered by the original code's license and performance.

However, the move has drawn sharp criticism. The library's original author, Mark Pilgrim, who wrote and released it under the LGPL in 2006, has objected on GitHub. Pilgrim contends the new version is a modification of his original work and therefore must retain the same LGPL license, not bypass its restrictions via an AI-assisted rewrite. The dispute highlights unresolved questions about how traditional software licensing applies when AI tools are deeply involved in recreating functionality, testing the legal and ethical boundaries of the 'clean room' engineering concept in a new era.

Source: Ars Technica