AI Trust and Safety: Lessons from OpenAI | encorp.ai