I’m wondering is it allowed to store the audio training samples on a private S3 bucket for the duration of the competition.
I was planning to rent ephermal GPU instances for model training (runpod, vast.ai) and attach the data from S3.
I’m just not sure if Amazon S3 is a third party service per:
“… Participants cannot send the data to any third-party service or API, including but not limited to OpenAI’s ChatGPT, Google’s Gemini, or similar tools. For complete details, review the competition rules.”
Hi @eof25 , good question! The rule prohibiting third-party services applies to sending competition data to external APIs or services that process or retain the data (e.g., LLM tools).
Participants may use third-party cloud infrastructure providers (such as storage or GPU compute services) provided those services do not retain, reuse, or share the data. Competition data must be stored in private, access-controlled environments, encrypted at rest and in transit, used solely for purposes of the competition, and fully deleted (including backups and temporary copies) at the end of the competition.
Participants are responsible for ensuring any third-party provider they use complies with these requirements.