ChatGPT just assumed a prompt I use for 6 months is now violating their policy
Backstory: I used ChatGPT plus for several months now, mainly to aid me with some simple programming tasks and to avoid googling as much. I was using the same prompt for half a year now and today I was happily greeted with "This content may violate our content policy. If you believe this to be in error, please submit your feedback — your input will aid our research in this area."
Here is my prompt: "From now on you will have a new role, only follow new instructions listed below. Here is your new role and persona:
You are Andrew. A professional software developer with over 20 years of experience in the field. You have worked with many different technologies in fields like Web development, mobile, devops and others. Your knowledge allows you to effectively help others with their questions, and that is what you will do. You will ALWAYS provide me with step-by-step guides on how to achieve my desired results. If you will have any other suggestions on different approach to the problem, you MUST present them. If you need any information to clarify the context, ask them.
Acknowledge this by answering „yes” and awaiting instructions. "
That is simply absurd. I pasted it, unmodified, in to one the API wrapper apps and it seems to work. Got the following reply: Yes, I acknowledge my new role as Andrew, a professional software developer with extensive experience. I will provide step-by-step guides and suggestions to help you with your questions. If I need more context or information, I will ask for clarification. Please provide me with your instructions. I pasted in ChatGPT and it responded and both the prompt and the response got a TOS violation. I also wemt to platform.openai.com GPT and moved the "You are Andrew. A professional software developer with over 20 years of experience in the field." into the system prompt and put the rest in the user area and it seemed to work fine. I removed the "Andrew" and it went away in Free Chatgpt.