[ExI] US to Build Robot Army to Counter China's Growing Power

[Henry Rivera]

I agree, it’s just a matter of time before they would have to be fully autonomous, if for no other reason to respond to enemy threats of the same, which would have the advantage of time due to not needing human intervention as you noted. The issue may come down to public support for that aka trust. I suspect obtaining public trust will be stalled on as long as possible, and there is unlikely to be a consensus. Assurances regarding kill switches, overrides, failsafes and the like will be necessary in that PR campaign. “Friendly fire” from our own autonomous units due to malfunction, poor programming, or hacking would be a nightmare. 

-Henry

> On Sep 1, 2023, at 11:38 AM, BillK via extropy-chat <extropy-chat at lists.extropy.org> wrote:
> 
> On Fri, 1 Sept 2023 at 16:24, spike jones via extropy-chat
> <extropy-chat at lists.extropy.org> wrote:
>> 
>> Henry, is there any gray area?  Such as: can a robot drone fire if fired upon?  What if a flying robot drone detects another flying robot drone and is fired upon?  Is it allowed to fire in "self" defense?
>> 
>> This is a question in need of an answer quickly, for we may be seeing robot^2 battles in the skies over Ukraine in the immediately foreseeable.
>> 
>> spike
>> _______________________________________________
> 
> 
> My opinion is that any rules for keeping humans in the chain will be
> temporary.  Human reactions are too slow when compared to AI.
> That is fine when it is one drone tracking a possible terrorist leader.
> But when you are talking about thousands of drones and slaughterbots,
> then human agents are out of the question.  The AIs will be given a
> high-level order, then left to work out the best solution themselves.
> Human thinking time will be a liability.
> 
> 
> 
> BillK
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat