Mapping the brain – or using imaging to see how the brain works -- is still in its infancy, and there is a long way to go before it's fully understood. However, it could very well be what brings us to the next level of artificial intelligence.
Eventually, we could see countless "human-like" tasks that can be transferred to robots. This topic is top-of-mind for Nir Shavit, CEO and co-founder of Neural Magic. He recently spoke to Avishai Sharlin, general manager of Amdocs Technology, about what could make this possible.
"I think deep learning is going to become a very large and commoditized area of our computer infrastructure. We're going to use these algorithms to improve the efficiencies of everything that we do because we have so much data, and we can't analyze it by hand anymore," said Shavit. "And there's no time for people to sit and invent individual algorithms for things. Over time, we will build systems that learn a simple subset of tasks and get better and better."
While super-intelligent robots on the level of a human brain could very well be in our future, it'll take some time. In the meantime, we'll see some gradual steps that help us steer our increasingly digital world.
"Right now, the way the digital economy works is that if I want a pair of shoes, I have to go through many websites with thousands of pairs of shoes until I find the right one. This is an example of where simple things, like simple recommendation systems and machine learning, can completely change the economy," said Shavit. "And it doesn't have to be human intelligence. It's just an algorithm to help you navigate this digital world. I think that's an example of what we're going to see long before seeing artificial intelligence and robots."
Listen to Shavit and Sharlin discuss this further on the Future of Tech podcast series.