I asked Copilot Researcher a question and it gave me an answer, but the longer you use it the worse it gets. Then I asked it to refine its answer and gave it a simple instruction. An hour passes and then it tells me, it couldn’t answer it. Do I want to do something else? I had to wait an hour to find out you couldn’t do it?
I asked it again and lets see what happens. With Copilot it feels less reliable than ChatGPT. Copilot Researcher can give you wonderful information but will it work? I’m shooting 50/50 with it when I ask it to do something. I get that it can have an unexpected problem, but honestly after waiting an hour to get nothing is not acceptable. Most people won’t use it again if they feel it wasted their time.
It is frustrating for me because I am using multiple AIs at once. I have ChatGPT queries going, and Copilot queries going in Teams/Outlook Copilot, and M365 Copilot app. At any time, I have at least 2 AI queries going. Why am I using so many? Because if it takes an hour or a half hour to answer a question, I have things that I need to do. Most people will only have access to 2 Copilots if they have Outlook/Teams so that limits its usefulness.
When I have tried to do a second query in the same Copilot it does one of two things. 1. It cancels the first query when you switch back to it. 2. It forgets the first query or has some unrecoverable error. Why is having more than one query a problem with an AI? If we are a customer, why aren’t we getting full functionality from it? Why are we being limited and have such poor error condition handling. Honestly, AI trips more than a drunk monkey in sneakers.
*breathes* AI has a long way to go to be useful.