The output isn’t interesting if we don’t know what the prompt was. What, specifically, concerns you about this output without knowing the prompt? > Brazen fal

The output isn’t interesting if we don’t know what the prompt was. What, specifically, concerns you about this output without knowing the prompt? > Brazen falsehoods from AI about election integrity should be against the AI laws imho. So AI should be forbidden from writing fiction about elections? …from generating an example of what a political demographic thinks for use in a discussion? Your comment here presumes that it’s spreading “brazen falsehoods”, instead of writing fiction or emulating a particular world view in response to an explicit request — but that’s exactly what we don’t know without the prompt (and why I called it misleading). Showing the output without the prompt is equivalent to Media Matters spreading screenshots without sharing their methodology to generate them.