RE: Resolved: Human plus AI will always outperform human alone and AI alone.
Humans have taught artificial intelligence to learn. Very quickly it will acquire more knowledge than a person, will gain some experience. Over time, AI will teach other AI. I am sure that humanity will still have some problems with this. Maybe not like in many fantasy films. But over time, AI will not only answer questions, it will certainly learn to act. It is not known exactly how this will happen. Perhaps people will create robots with AI, or AI will manage, for example, technological processes in factories. The more complex the system, the more likely it is to fail or produce incorrect results. AI will definitely come to certain wrong conclusions at some point, and in accordance with them, it will change the process that it will control.
A lot of that stuff is already happening now, especially with factory and farm robotics.
Certainly, there will be risks and harms like the ones that you describe, but I hope they will be limited by the human owners and operators of the AIs and by market and political competition that will eliminate the most dangerous systems.
Historically, new technologies have always created new risks and new rewards, and I expect that to continue.