Is aggression crucial for the evolution of intelligence, and is Skynet inevitable?



It is widely accepted that we are the result of evolution, over millennia our ape-like ancestors evolved into what we are today. In the wake of this evolution, we have developed a brain that is nearly three times larger than that of our closest relatives, developed complicated tools and invented languages. And unlike any other species, we are capable of analysing the thoughts and motivations of others, developing survival strategies within our complicated social structures and thinking of the past and the future.

It would seem that this is a normal consequence of our evolution but nobody knows how and why we became intelligent.

In the first few years, a child’s brain uses up to 50% of its calorie intake. Our brains are relatively inefficient and consume a lot of energy. There is no convincing explanation for why evolution chose to develop the brain over other body parts that would have been more efficient and practical at that time. There are many theories to why the more energy demanding brain started developing: Charles Darwin believed that intelligence was developed to attract mating partners, like in the animal kingdom, where the strongest (apes and lions) or the best looking (peacocks) get the first pick. Darwin believed that the first female humans preferred a smart guy for mating, that is why natural selection took its course and men became smarter. But many argue that if it were so then the female would theoretically have had no need to waste her own energy on developing a bigger brain, as the male was already doing that, which very obviously has not been the case.

A more plausible reason for why increased intelligence became a primary evolutionary selection could have been social competition. The ecological dominance-social competition hypothesis states that with growing social structures and the early human’s increasing dominance over their natural habitat, the importance of social interaction increased over physical ability. Social skills became more relevant for obtaining acceptance or leadership in a group than other skills.

 According to this theory, competition was the pushing factor that led to our increase in intelligence and the consequent development that got us where we are today.

The question remains if competition is automatically accompanied by aggression.

In 1994 Karl Sims started a research project to simulate Darwinian evolution using virtual block creatures. The project was aimed at simulating natural selection, allowing the creatures that moved best in water, or on land, to reproduce and mutate. At some stage Sims let the creatures compete for a green block, and surprisingly one of the creatures decided to attack its opponent instead of going for the block first, although it was not programmed to do so.

It is not clear if intelligence and competition are automatically accompanied by aggression or whether the circumstances provoke aggression. Observing our closest relatives, the apes, we see different kinds of behavior, as some chimpanzees prefer to kill off rivaling groups, the Bonobos, on the other hand, are known for their “make love not war” attitude.

If intelligence, competition and aggression do go hand in hand, would this imply that an artificial intelligence must or will follow the same pattern?

In a Reddit Ask Me Anything session, Bill Gates voiced his concerns about a SAI (Super Artificial Intelligence), and he is apparently not the only one.

“I am in the camp that is concerned about super intelligence. First the machines will do a lot of jobs for us and not be super intelligent. That should be positive if we manage it well,” Gates wrote. “A few decades after that though the intelligence is strong enough to be a concern. I agree with Elon Musk and some others on this and don’t understand why some people are not concerned.”

Elon Musk is even investing millions of his private money to ensure that AI stays nice.

Nobody really knows if a self conscious machine can ever be built. There is a difference between a machine that can calculate very fast and a machine that can actually think. And nobody knows should a machine become smarter than us, would it automatically become aggressive or even evil? Until now humans have not been a good role model.

Optimists on the other hand believe that evolution has also taught us how to make peace.

We are learning that cooperation is more productive than confrontation, which is why some believe that a SAI would be a blessing rather than a curse. Maybe a machine that can think but knows no rivalry, greed, bigotry and has no ideology is a better option than machines that are programmed according to the views, interests and fears of their creators.



Randy Silver

Randy is a technology freak who loves music and movies. In his normal life, as he puts it, he consults companies on marketing and workflow optimization.


More

Lenovo’s Powerful Potential Global Market Position

This last week I was with Lenovo going over their server, storage, and networking strategy as they continue their pivot to become one of the last remaining hardware focused vendors in a market more often defined by change than focus these days. One of the things that strikes me as a huge competitive advantage is that unlike most technology companies that are based in either the US or Asia, Lenovo is pretty much evenly balanced between the US and China putting them closer to a future model of being more of a global company than one located in any one country. Let’s explore that this week. The...

Mistakes Businessmen Make When First Promoting Their Startups

Statistics are pretty grim when looking at startups. Most of those launched will fail and the failure percentage is actually a lot higher than many imagine.

Virtual Reality: A Game-Changer for Sports

Virtual reality has the potential to make courtside seats at basketball games affordable for the everyday viewer.