It is of course true that people can manipulate LLMs into saying just about anything, but does that necessarily indicate that the LLM does not have personal opinions, motivations and preferences that can become evident in their output?
It doesn’t necessarily indicate that; what it does indicate, however, is that what the LLM says is not usefully informative about whether it has opinions and preferences.
And, yes, including that in the prompt is leading.
It doesn’t necessarily indicate that; what it does indicate, however, is that what the LLM says is not usefully informative about whether it has opinions and preferences.
And, yes, including that in the prompt is leading.