AerynOS, an alpha-stage Linux distribution, has implemented a policy banning large language models in its development and community activities. The move addresses ethical issues with training data, environmental impacts, and quality risks. Exceptions are limited to translation and accessibility needs.
AerynOS, a Linux distribution focused on atomic updates and still in its alpha phase, has updated its contributing guidelines to prohibit the use of large language models (LLMs) throughout the project. This decision, announced on Reddit, applies to all aspects of development and community engagement, including source code, documentation, issue reports, and artwork.
The policy stems from several key concerns. Developers highlight ethical problems with how LLMs are trained, including the sourcing of data. They also point to the high environmental costs, such as excessive electricity and water consumption involved in building and operating these models. Additionally, there are worries about how LLM-generated content could degrade the overall quality of contributions and raise potential copyright issues.
While the ban is comprehensive, AerynOS allows narrow exceptions. Contributors can use LLMs only to translate text into English for issues or comments. The project may consider further allowances for accessibility purposes. In terms of user support, the team advises against depending on AI chatbots over official documentation. Requests based on inaccurate LLM outputs risk being overlooked, as maintainers aim to avoid debugging third-party errors.
This forward-looking policy seeks to ensure that all contributions undergo human review, thereby upholding the project's technical standards and reliability. It reflects a growing trend among open-source initiatives to scrutinize AI integration amid broader debates on its implications.