In a recent decision by the Gentoo Linux distribution council, a ban on the use of code created using artificial intelligence has been announced.
This decision comes after Mikhala Gurny, a member of the council, raised concerns about the risks of copyright issues, quality control problems, and ethical considerations associated with the high energy consumption of AI and the influence of large corporations on technology development. Gurny highlighted that AI models trained on protected materials can pose copyright problems, generate meaningless text and code, and even “see” non-existent software packages. The Gentoo community, known for its traditional approach to program engineering, where emphasis is placed on humans rather than productivity, is expected to view this ban positively.
The discussion surrounding the ban took place in e-mail and IRC chats, with a consensus emerging on the need for some restrictions. The ban was officially approved at a council meeting on April 14, with six votes in favor and one member absent.
Implementing the ban may prove challenging as distinguishing between human-written code and code generated by AI is not always straightforward. The council’s main objective is to clearly communicate the rules to participants in a polite manner. The policy allows for a review in the future, as advancements in technology could impact decision-making. The council is exploring the possibility of exceptions for AI trained specifically on Gentoo materials to avoid copyright issues and potentially enhance code quality.