Someone once asked...

“In recent regulatory proposals, tech companies suggested an AI safety threshold: a model only needs to be halted if it poses a risk of causing 50 or more deaths. This raises a chilling question—can human lives really be assigned a numerical quota?" "For the sake of technological progress, can we ever accept a small, predictable cost in human lives—or is quantizing life a line we should never cross?”

✦

Someone answered with kindness...

It is a line we should never cross...but how on earth do you predict loss of life in this scenario? If it's a significant enough chance that it's worth measuring, that's an issue period.

✦

Wandering voices — answers no one has read yet

✦

Another soul answered...

I think the tragic answer is yes to both, technological advancement is built on a foundation of bodies, it always has been, but I also believe that we should never disregard the sanctity of human life. If we can save lives, we always should, and I personally believe that AI systems will never truly advance humanity, and will always have a cost to human life. I think the entirety of the system should be disregarded and dismantled.