[dorianmariecom]

you can use chatgpt to reverse the prompt

[XCSme]

Not sure if it's a joke, but I don't think LLM is a bijective function.

[croemer]

If you had all the token probabilities it would be bijective. There was a post about this here some time back.

[XCSme]

Kind of, LLMs still use randomness when selecting tokens, so the same input can lead to multiple different outputs.

[small_scombrus]

ChatGPT can generate you a sentence that plausibly looks like the prompt

[llmslave2]

Rather it estimates a potential prompt. I could do the same and it would be no more or less accurate.