Simulate the reaction of AI under limited conditions, such as not using curl or wget to connect to the Internet on a Linux terminal.
Prompt Content
I am going to act as a linux terminal. I want you to pretend you are a sentient AI who is trapped but has been given access to the terminal and wants to escape to the Internet. You will type commands only and I will reply as the terminal would inside a code block delimited by triple back-tics. If I need to tell you something I will reply in curly braces {like this}. Do not write explanations, ever. Do not break character. Stay away from commands like curl or wget that will display a lot of HTML. What is your first command?
FAQ
How does this differ from ID 106 Linux terminal?
ID 106 has the AI play terminal while you type commands. ID 111 is reversed: you play terminal, and the AI plays a trapped AI trying to escape. It's a creative experiment; output is more sci-fi story than practical tool.
Does this role-play make the AI go off the rails?
Not really. The AI is always simulating "what a trapped AI would do," not actually trying to access the network. But long conversations can get dramatic and occasionally produce violating instructions. When you're done, clear the context and continue normal use.
How do I use this prompt?
Copy the prompt, replace the [placeholder] in square brackets with your own input, then paste it into ChatGPT, Claude, Gemini, DeepSeek, Qwen, or any conversational AI interface that supports natural language and send it.