policy on AI responses?

[permafrost]

Does PAN have a policy, a community rule with respect to copy-pasting of AI responses when responding to questions?

If not - no restriction on drive-by AI-generated walls of text (more often irrelevant, sloppy, or even plain wrong - than not)?

(To my 👀, one of the biggest challenges with both AI and human responses is insufficient context - a challenge especially applicable to newbies like me who often just don't know what context is relevant or important and need to be guided through it. The difference between humans and most LLM implementations is that thinking humans ask for missing context or to rephrase a question whereas LLMs press on with walls of texts that presume missing context - and more often than not get it wrong. ... and by then the thread is polluted and becomes unreadable.)

[JayGolf]

Hi @permafrost ,

We actually do have an AI Guideline, you can check it out here.

The gist of it is that AI can be used as a writing or productivity aid, but it should NEVER replace real technical understanding, validation, or hands-on experience. Copy/pasting raw AI-generated answers into threads is not considered acceptable, especially when the answer has not been reviewed or verified.

The way I personally like to think about it is: AI should augment someone’s thinking, not replace it. In our case, that means the final answer still needs to be grounded in personal technical experience. Which is think is the most important part when we are trying to help fellow humans solve real problems. 

[permafrost]

This is great, thank you!

Then the next questions are about detection, reporting and enforcement:

• I am guessing if I see a response that looks like copy-paste AI slop, I report it?
  • If it doesn't get taken down - like here, here and here - and I get no updates on my reports - what's next?
• Given that prolific, widespread use of AI slop in this community is unlikely to be stopped by manual reporting and enforcement - does Palo Alto plan on implementing any automated mechanisms?

Thanks again!