Preempting the Risks of Generative AI: Responsible Best Practices for Open-Source AI Initiatives