I have yet to hear a convincing scenario for how AI kills a significant portion of humans in 10 years. Getting AI to “align” to human judgment and values is never going to work completely, humans judgment and values are contradictory tradeoffs for a single human and no two humans see eye to eye on everything. Beyond that, achieving “alignment” is something we have failed to achieve and repeatedly failed to achieve with humans each and every time we make a new one. Why would we be able to achieve it with AI? Our prisons are filled with humans that refused to align to a base sort of defined criteria. How can it be that they haven’t sat and thought about this problem for more than a few weeks and come to the same conclusion, at this point it’s almost like these companies must be filled with super technical young engineers that lack experience with and a complete grasp of what it means to be human to start with… oh