Sharing Is Caring, But So Is Quality

[rubi1200]

Dear Community,

I want to share a blog post that I just came across over at Meta Discourse.
https://meta.discourse.org/t/ai-can-code-but-it-doesn-t-care-about-quality/392913

The post discusses the usage of AI for software coding but there are some very relevant lessons we can also learn for our Discourse.

Here are the key takeaways that I see from a non-tech perspective:

• AI is here to stay, I think we all know that. We should embrace it but be very cautious and selective when using it
• AI has its uses, like any other tool. But it also has its limitations. Never underestimate the potential damage from even the smallest error in a command or script
• Cross-check, validate, confirm before you let code loose in a post. Test the code or script in a VM or on a test install. Please don’t accept code or script from an LLM without doing due diligence
• We all want to help our fellow community members but we also don’t want to have to clean up the damage if something goes wrong

Finally, a reminder to everyone of our guidelines regarding AI in posts:

@system

Your answer should be yours and not just something you have regurgitated from the internet. By this we mean that you have tried it (or something very similar) and you have a good idea of what it will do, won’t do, and when it might not be appropriate. This includes using search engines or AI to help you come up with an answer.

• If you know how to make good use of AI and are sure that your result is valid, you may use it in helping you to formulate your answer, but please acknowledge that you have used it. Any posts which turn out to be entirely or largely AI, especially without acknowledgement, will be removed and infractions may result.

My emphasis added.

[WyattWhiteEagle]

Thanks for posting this, very helpful.

I’ve noticed that web browser search engine’s are including AI generated search results in their results output.

I don’t believe they can be fully trusted as “best solutions” but they can be very helpful.

What you share here is very informative.

Something I notice in AI result suggestions is the link saying “mistakes may be included”.
User’s need to pay attention to what is mentioned at those pages which are linked to.

Please fill in the template here at discourse mentioning what all you have tried.
It really help’s those who are trying to help you.