My question is what safeguards have you put in place not to get a huge bill from OpenAI in case users/bots abuse your interface. Assuming you're using the API a

My question is what safeguards have you put in place not to get a huge bill from OpenAI in case users/bots abuse your interface. Assuming you're using the API and not an alternative GPT model like GPT-J.