EmbeddingGemma 2: an open, lightweight multimodal embedding model
deepmind.google·6d ago
AI Engineering
Lobsters·Jul 23·5 min read
Protecting our FLOSS commons from LLMs
TL;DR
Codeberg e. V. members voted to prohibit the use of their data for training Large Language Models (LLMs), citing concerns over the impact of LLMs on the free/libre software ecosystem.
✦ Why It Matters
Consider reviewing your data usage policies to align with community standards on AI and data privacy.
Key Takeaways
- Review your project's compliance with Codeberg's new Terms of Use.
- Avoid using LLMs for project development to align with community standards.
- Consider the environmental impact of LLMs on digital infrastructure.
- Engage in genuine collaboration to enhance the free/libre software community.
Full Summary
Codeberg e. V. held a vote among its members regarding the use of data for training Large Language Models (LLMs) and the impact of 'vibe-coded' projects.
The organization reaffirmed its commitment to not using user data for LLM training, aligning with its privacy policy. Additionally, a controversial motion to change the Terms of Use to prohibit 'vibe-coded' projects was accepted, reflecting concerns over resource consumption and the integrity of the free/libre software community.
The rising costs associated with LLMs, including hardware and energy, threaten the sustainability of digital infrastructure. Codeberg aims to foster genuine collaboration and maintain a healthy ecosystem by discouraging projects that do not contribute to community-driven development.