Professor Stuart Russell explained that Hollywood’s depiction of killer Artificial Intelligence robots misrepresented what the biggest concern should be. While on the Today Programme Mr Russell explained that scientists and engineers designed AI with goal-orientated mindsets. This, in turn, means that Artificial Intelligence programmes will choose the easiest and simple methods to achieve their goals which could mean killing or eliminating humans. In order to stop robots potentially destroying the human race, Professor Russell argued scientists could change the way AI is engineered so it doesn’t focus on a goal-orientated mindset.
Mr Russell insisted that unless this changed, AI could lead to human extinction as we give it the tasks of solving complicated problems like crime and climate change.
He said: “If you think about it, the way we build Artificial Intelligence is a bit like the way we think of a genie in a lamp.
“You rub the lamp and the genie comes out and you say I want this to happen.
“If the AI system is sufficiently powerful it will do exactly what you ask and you will get exactly what you asked for.
“Now the problem with genies in lamps is the third wish is always to undo the first two wishes.
DONT MISS: What NASA got wrong about God of Chaos Apophis asteroid exposed
“This is because we are unable to specify the objectives correctly.
“So a machine pursuing an objective that isn’t the right one becomes an enemy of the human race, an enemy that is much more powerful than us.
“So if you are going to build things that are much more powerful than on Earth do you make sure that they never have power over you and that is the question that we are trying to answer.”
The Today Programme host then asked for an example of Artificial Intelligence belong an enemy of the human race.
Mr Russell replied: “I think the Hollywood meme is always that a machine spontaneously becomes conscious and then, of course, decides that it hates human beings and wants to kill us all.
“That is completely the wrong point to be worrying about.
“So it is not really evil consciousness that we have to worry about it is competence.
“Just competence in pursuing an objective that is misspecified by us.
READ MORE:
Antarctica: Scientists make breakthrough over dinosaur-extinction [VIDEO]
NASA asteroid revelation: Space rock ‘threatens’ Earth – researcher [ANALYSIS]
Asteroid tsunami: Why scientist offered dire warning to US coast [COMMENT]
“For example, if we said we have a very intelligent climate control system at some point in the future and we want to return carbon dioxide levels to pre-industrial concentrations so the climate can get back into balance.
“The AI system then figures out the easiest way is to get rid of all human beings because they are the ones producing all of the carbon dioxides in the first place.
“Then you might say, okay well you could programme so that the AI can do anything it wants but you cannot harm human beings.
“So what the system will do is just convince us to have fewer children until there are no human beings left.”
Source: Read Full Article