4 ms·
More effective warfare in the right hands would kill less people. We're fine when America keeps the military upper hand, we're fucked if nations who care less a
by lemagedurage 8y ago
More effective warfare in the right hands would kill less people. We're fine when America keeps the military upper hand, we're fucked if nations who care less about keeping people alive do decide that using neural nets for warfare is a sound idea.
- dannypgh 8y agohttps://en.wikipedia.org/wiki/List_of_countries_by_military_expenditures https://en.wikipedia.org/wiki/List_of_countries_by_military_... https://en.wikipedia.org/wiki/Arms_industry#World's_largest_arms_exporters https://en.wikipedia.org/wiki/Arms_industry#World's_largest_... Do you think this technology will really "disrupt" which nation is both the largest manufacturer and consumer of weapons? I think who the "right hands" are to handle military power is something that reasonable people can (and do) disagree on...
- Sol- 8y ago>More effective warfare in the right hands would kill less people Debatable. Putting too much trust into technological solutions to make the killing decisions for you may as well reduce oversight and transparency. After all, if the algorithm said that the guy was a terrorist and deserved to die, who would question it? It's supposedly an unbiased machine making the call, right? Algorithmic transparency is really not understood enough to even consider autonomous warfare. If Google wants to pretend that it's doing something ethical when working in that space, most of their focus should be on that instead of making drones kill better.
- simion314 8y agoI personally would not want to work on killer automatic robots even if is a challenging domain and some other country is doing it, if someone else wants to work on that then I am ok and a society we need to find a way to make this things illegal(like we have done for land mines,chemical weapons etc)
- jakobegger 8y agoThe US already kills a lot of people with drone strikes based on highly questionable intelligence. I’m not sure how involving “AI” would make the process of deciding who to kill any less arbitrary.