AI Rewrite Sparks Licensing War Can Chardet Shed its LGPL Roots for an MIT Future?

 

The Chardet Controversy: AI-Generated Code, Licensing Wars, and the Ghost of Google v. Oracle

A heated dispute has erupted within the Python community involving Mark Pilgrim, the original creator of the chardet library, and the project’s current maintainers. The conflict centers on the release of version 7.0.0, which saw a dramatic licensing shift from LGPL (Lesser General Public License) to the more permissive MIT License. The justification for this change? The current team claims the entire codebase was rewritten from scratch using Artificial Intelligence (AI).

The Legal Battlefield

Mark Pilgrim has expressed strong dissatisfaction, demanding a reversion to the original LGPL license. This has sparked a complex legal debate on several fronts:

• API and Fair Use: Under the U.S. Supreme Court’s ruling in Google v. Oracle, recreating software with identical APIs is generally protected as Fair Use.

• Derivative Work vs. Transformation: Critics argue that using AI to rewrite code could be viewed as a mere "automated translation" or "tool-assisted conversion," making the output a derivative work of the original LGPL code, thus legally bound by the original license.

• AI and Copyrightability: U.S. federal courts and the Copyright Office have previously ruled that AI-generated works lack human authorship and are therefore not eligible for copyright protection unless there is "sufficient human expressive contribution."

The Maintainer’s Defense

Dan Blanchard, the current lead maintainer, defended the move by clarifying that the AI (Claude) was guided solely by public API specifications and test datasets not the original source code. Blanchard emphasized his "human-in-the-loop" oversight and noted that version 7.0.0 bears less than a 1.29% structural similarity to previous versions.

The ultimate goal for this re-licensing is to pave the way for chardet to be officially merged into the Python Standard Library, which strictly requires an MIT or BSD-compatible license.

If Blanchard's work gains widespread legal acceptance, it could become a new norm known as "license laundering," or the process of AI rewriting copyleft code (such as the GPL) to create a different license. This could shake the foundations of freedom in the open-source world, which emphasizes sharing and giving back to the community.

The 1.29% figure, while seemingly small in quantity, is more important in copyright law than quantity. If the identical parts represent the "core" or "fundamental logic" that Pilgrim invented, the court might still consider it a derivative work. This will be a crucial test of how AI can overcome the limitations of derivative works.

The attempt to incorporate the library into the Python Standard Library is a significant undertaking (similar to the case of requests in the past). The team's choice to perform a "major overhaul" using AI reflects the fact that past license issues were a major bottleneck hindering software growth. They view AI as the quickest solution to legacy legal debt.

If this case ends with acceptance of the AI ​​code, giant tech companies will be able to use AI to "reverse-engineer" open-source software for internal use without worrying about disclosing the source code under the original GPL agreement, which would forever change the face of competition in the software industry. 

 

Why OpenAI Skipped GPT-5.3 Everything You Need to Know About the New GPT-5.4.

 

Source - GitHub: chardet/chardet

💬 AI Content Assistant

Ask me anything about this article. No data is stored for your question.

 

The Chardet Controversy: AI-Generated Code, Licensing Wars, and the Ghost of Google v. Oracle

A heated dispute has erupted within the Python community involving Mark Pilgrim, the original creator of the chardet library, and the project’s current maintainers. The conflict centers on the release of version 7.0.0, which saw a dramatic licensing shift from LGPL (Lesser General Public License) to the more permissive MIT License. The justification for this change? The current team claims the entire codebase was rewritten from scratch using Artificial Intelligence (AI).

The Legal Battlefield

Mark Pilgrim has expressed strong dissatisfaction, demanding a reversion to the original LGPL license. This has sparked a complex legal debate on several fronts:

• API and Fair Use: Under the U.S. Supreme Court’s ruling in Google v. Oracle, recreating software with identical APIs is generally protected as Fair Use.

• Derivative Work vs. Transformation: Critics argue that using AI to rewrite code could be viewed as a mere "automated translation" or "tool-assisted conversion," making the output a derivative work of the original LGPL code, thus legally bound by the original license.

• AI and Copyrightability: U.S. federal courts and the Copyright Office have previously ruled that AI-generated works lack human authorship and are therefore not eligible for copyright protection unless there is "sufficient human expressive contribution."

The Maintainer’s Defense

Dan Blanchard, the current lead maintainer, defended the move by clarifying that the AI (Claude) was guided solely by public API specifications and test datasets not the original source code. Blanchard emphasized his "human-in-the-loop" oversight and noted that version 7.0.0 bears less than a 1.29% structural similarity to previous versions.

The ultimate goal for this re-licensing is to pave the way for chardet to be officially merged into the Python Standard Library, which strictly requires an MIT or BSD-compatible license.

If Blanchard's work gains widespread legal acceptance, it could become a new norm known as "license laundering," or the process of AI rewriting copyleft code (such as the GPL) to create a different license. This could shake the foundations of freedom in the open-source world, which emphasizes sharing and giving back to the community.

The 1.29% figure, while seemingly small in quantity, is more important in copyright law than quantity. If the identical parts represent the "core" or "fundamental logic" that Pilgrim invented, the court might still consider it a derivative work. This will be a crucial test of how AI can overcome the limitations of derivative works.

The attempt to incorporate the library into the Python Standard Library is a significant undertaking (similar to the case of requests in the past). The team's choice to perform a "major overhaul" using AI reflects the fact that past license issues were a major bottleneck hindering software growth. They view AI as the quickest solution to legacy legal debt.

If this case ends with acceptance of the AI ​​code, giant tech companies will be able to use AI to "reverse-engineer" open-source software for internal use without worrying about disclosing the source code under the original GPL agreement, which would forever change the face of competition in the software industry. 

 

Why OpenAI Skipped GPT-5.3 Everything You Need to Know About the New GPT-5.4.

 

Source - GitHub: chardet/chardet