Toby Walsh,
according to his personal bio page,
“is Research Group Leader at NICTA. He is adjunct Professor at the Department of
Computer Science and Engineering at the University of New South Wales, external
Professor of the Department of Information Science at Uppsala University and an
honorary fellow of the School of Informatics at Edinburgh University.”
As of
October 8, 2015, his “How can you stop killer robots?” TEDx Berlin talk has
been available online here.
Recently,
Russian television “inadvertently” broadcast plans to build a super-secret
autonomous submarine capable of “delivering” nuclear torpedoes off the coast of
potential adversaries. Read about that
here.
Following
up Professor Walsh’s TEDx talk and news of this Russian initiative to build an
autonomous submarine with nuclear capability, Etopia News contacted
Professor Walsh by e-mail, and asked for his views on this latest development
of a potentially-lethal mega-killer robot.
Within the hour, here’s what he thoughtfully had to say:
“Couple
of comments.
“First, this demonstrates that the sort of autonomous weapon technologies we've been talking about are, as we warned, very near.
“Second, we don't have to worry about non-state actors or terrorists with these sort weapons. You still need a nuclear device which is beyond the means of such groups.
“Third, I'd be very concerned about control of such weapons ... once launched, there is likely no way of changing your mind and preventing them from attacking, and I'd be concerned about them being "hacked".
“Fourth, this is, again as we warned, likely to start an underwater arms race as these are very difficult technologies to defend against.”
“First, this demonstrates that the sort of autonomous weapon technologies we've been talking about are, as we warned, very near.
“Second, we don't have to worry about non-state actors or terrorists with these sort weapons. You still need a nuclear device which is beyond the means of such groups.
“Third, I'd be very concerned about control of such weapons ... once launched, there is likely no way of changing your mind and preventing them from attacking, and I'd be concerned about them being "hacked".
“Fourth, this is, again as we warned, likely to start an underwater arms race as these are very difficult technologies to defend against.”
Of course,
as you can read here,
DARPA is already working on autonomous naval systems to track submarines,
including autonomous ones, leading to a situation in which “finally, we may
find our submarines themselves going unmanned one day as well, roaming the
depths for months on end, looking for and hunting other unmanned and manned
submarines. It's the ultimate robotic game of cat and mouse, with patrols
lasting years and the ability to track other subs for months on end.”
The
struggle for military superiority thereby devolves into a competition between
rival teams of coders developing the learning algorithms that these fleets of
prowling, autonomous craft will use to optimize their performance in protecting
the geo-strategic and tactical interests of their operators.
2 comments:
"We can't afford a doomsday gap, Mr President!!'
"We can't afford a doomsday gap, Mr President!!'
Post a Comment