[RajT88]

I am in this boat. My employer tracks daily copilot usage, and has gamified it across teams.

I ask SME's questions, and I get AI generated responses. The message it sends is this: don't bother me; I can't be bothered, just use AI. People are sending these low-effort Copilot writeups to our customers too - I can't imagine what they think of it.

The reason we pay SME's is because they can be trusted to provide correct answers that we don't have to be skeptical about. The people doing this haven't figured out yet they are undermining their own credibility and optics on willingness to help out the team by doing so.

Just this week, I had several people suggest, "Hey, have you asked copilot about [esoteric networking thing]?". Indeed, I have, and in the absence of documentation, it gave me 5 convincing theories - none of which actually checked out when I dug into them. It just made wrong shit up.

Most frustrating is the integration with our tenant. I try and ask questions about things I need deeper information on.

Me: Hey Copilot, can you dig me up more information on X thing?

Copilot: Have you considered (waves hand in sweeping motion), everything you yourself have written on the topic?

If I want real answers I turn to the free ChatGPT, or my personally paid Claude subscription. Then I copy-pasta the stuff which is useful, and maintain my own writing style, and present the information in a way which I think works best for who I am talking to.

[AlexeyBelov]

Have you tried calling them out on it? I mean, use AI all you want, but dialogue with your coworkers still needs to be respectable. Pasting a "let me google that for you" link was always a no-no.