I also recently had to do a training module on AI. It was something my entire institution had to do. I work in the medical field, and I just... it was honestly horrifying to do that module.
The big focus was that AI could be such a great help, but we needed to make sure that we checked our work! Because we were responsible for anything we decide to use AI for. And in my mind, I'm like: "...ok, great. I will take full responsibility for anything I use AI for. I will not use AI at all."
Besides all of the existing issues with it (theft of intellectual property to train the models, pollution, climate change, cost to the everyday person, the inaccuracies, etc), the main thing the module was trying to sell me on was was something I could already do. In the time it would take to look over something generated for notation, for example, I could have done it myself.
In a previous job, we'd made templates to copy/paste each time we had a call, and blank spots (usually with ____ ) to easily double-click to fill in the actual information. In a job that was already watching my speed per call, having to proofread/edit what GenAI makes would take me longer than just writing it myself.
Also there was a whole section in the module about how AI could hallucinate (they actually used hallucinate!), and I had one of those profound moments where I wondered if I was living in a simulation. Because it was so insanely stupid that I'm being told it's a good idea to use something that might just hallucinate data that doesn't even exist.
I feel like I'm living in an episode of Captain Planet. So cartoonishly evil and ridiculous but somehow is real.