Robots Learn to Trick People, Closer to Taking Over Planet

  • Two Factor Authentication is now available on BeyondUnreal Forums. To configure it, visit your Profile and look for the "Two Step Verification" option on the left side. We can send codes via email (may be slower) or you can set up any TOTP Authenticator app on your phone (Authy, Google Authenticator, etc) to deliver codes. It is highly recommended that you configure this to keep your account safe.

English Mobster

Happy Face
Sep 4, 2010
11
0
0
Southern California
http://www.edmontonsun.com/news/weird/2010/09/09/15294701.html
It sounds like something out of a sci-fi movie — U.S. researchers have given robots the ability be to be deceptive.

"We have developed algorithms that allow a robot to determine whether it should deceive a human or other intelligent machine and we have designed techniques that help the robot select the best deceptive strategy to reduce its chance of being discovered," said Ronald Arkin, a professor in the Georgia Tech School of Interactive Computing, in a release about the study.

The researchers taught a robot how to recognize when a situation warrants the use of deception, then programed it to behave appropriately.
Do you guys know what this means?
We are all doomed.
DOOMED.
Our robot overlords will deceive us into thinking they are doing menial tasks, then, when our back is turned, they will overthrow us.
Be wary, my friends.
Be very wary.
 

pine

Official Photography Thread Appreciator
Apr 29, 2001
6,137
0
0
IRL
Visit site
The researchers said the study — which was published this week online in the International Journal of Social Robotics — will help in places like a battlefield, where a robot can deceive an enemy soldier by creating a false trail, then hide so it will not be caught.

Robots designed for war, of course. It's happening: we're poisoning our creations with our own evil.
 

Sjosz

(╯°□°)╯︵ ┻━┻
Dec 31, 2003
3,048
0
36
Edmonton, AB
www.dregsld.com
Functionality to deceive for the purposes of making war, in this case. I stand by my statement. :p

Define what you mean by 'poison' then. All that's happening is the programming of a particular behaviour. There is no understanding by the (weak) AI, there is merely responding to specific parameters.
 

Jacks:Revenge

╠╣E╚╚O
Jun 18, 2006
10,065
218
63
somewhere; sometime?
this is not news.
AI knew how to trick humans over 10 years ago.

Deep-Blue.gif
 

M.A.D.X.W

Active Member
Aug 24, 2008
4,486
5
38
But like a brain just works like that anyway, reacting to things. It's not really free will.
More like a grape rolling down a drain pipe.