I don't disagree that most of the warnings perpetuated are probably part of a publicity stunt (or hype men). However there is a fair chance we will reach super

I don't disagree that most of the warnings perpetuated are probably part of a publicity stunt (or hype men). However there is a fair chance we will reach super-human autonomous actors in the next 50-100 years, and if we fail to align them very closely to human values it would be an extreme risk to all life. The point is that currently there is effectively zero effort put into (AI Safety research) an existential risk we can actually see coming. It may not happen at all (we could stop progressing compute efficiency/density before then) but trajectories suggest it will eventually. For a more nuanced discussion of the actual risks I recommend watching/listening to the ML Street Talk w/ Robert Miles: https://www.youtube.com/watch?v=kMLKbhY0ji0 https://www.youtube.com/watch?v=kMLKbhY0ji0