Donate your coding sessions to an open CC-BY-4.0 dataset to help train open-weight and open source models

A new initiative called Trace Commons is being launched to collect coding session data under a CC-BY-4.0 license, aimed at creating an open dataset for training open-weight and open-source models. This effort seeks to democratize access to training data, countering the dominance of proprietary models like Claude Code and Codex from Anthropic and OpenAI. By contributing coding agent traces, practitioners can support the development of more diverse AI models, fostering innovation in the open-source community.