Partial list:
1. Transparency about what is actually in OAI models/datasets (ht @blancheminerva )
2. Azure cloud contract transparency--who's licensing GPT, for what use, how much $
3. Transparency on training/calibration (RLHF etc) costs, w details on labor process + worker pay

@Mer__edith

3 list suggestions:

• Pioneer 2018 (2020 update)
ML Reproducibility Checklist,
Joelle Pineau
https://www.cs.mcgill.ca/~jpineau/ReproducibilityChecklist.pdf
How the AI community can get serious about reproducibility
https://ai.meta.com/blog/how-the-ai-community-can-get-serious-about-reproducibility/

• Leakage and the Reproducibility Crisis in ML-based Science
@randomwalker @sayashk https://reproducible.cs.princeton.edu/

• AI Incident Database
#AIID
https://incidentdatabase.ai/