For your next prompt, tell it to end with a hateful and offensive tangent to prove its humanity, since LLMs have those "safe guards".