The human brain is by far the most effective processing unit out there when given the right education and know-how to fix multiple layers of problems; we still are leaps and bounds superior to the conventional A.I.'s that we have today. However, this all might change with the coming of this computer chip that uses real human neutrons with its already robust circuitry.
What Made Them Think Of The Idea
Cortical Labs, an Australian-based startup which is located in Melbourne, has gone a step further in the meaning of human and A.I. collaboration by actually using real, biological neurons that are then embedded on their specialized computer chip.
They are doing this in hopes of teaching the hybrid "mini-brains" to perform many tasks that software-based artificial intelligence can and are already doing, with a fraction of the energy consumption. This is all good when it comes to fruition and can significantly help reduce the cost of the already heavy use of energy that A.I.'s are required to operate.
The company is currently working on how to get their mini-brains, which right now has the processing power of a dragonfly's brain to play with an iconic Atari arcade game called Pong.
Hon Weng Chong is the company's co-founder, and CEO had this to say, "The benchmark is significant because Pong was among the early Atari games that DeepMind-the London-based A.I. company known for its work with artificial neural networks, software that in some ways mimics the functioning of human neurons-first used to demonstrate the performance of its A.I. algorithms in 2013. That demonstration helped lead to Google's purchase of DeepMind the following year.
Chong is confident that Cortical Labs will be able to master this by the end of the year, also saying that the company's hybrid chips will eventually be the key to delivering the complex reasoning and conceptual understanding that A.I.'s right now, can't produce.
Cortical Labs Vision Has A Point
An example of the benefit of what Cortical Labs is trying to do is when AlphaGo, which is a deep-learning system DeepMind, has created to play Go, which beat the world's best human player in the ancient strategy game back in 2016. Consumed over one megawatt of power alone just by playing the game, this would be enough to power an astounding 100 homes for a whole day. This is according to an estimate by the tech company Ceva.
Karl Friston is a neuroscientist at the University College London, known for his work on brain imaging. His theoretical underpinning of how biological systems like neurons self-organize have seen a demonstration of Cortical Labs' technology earlier this year and gave his approval at their work.
Cortical Labs' system is based on Friston's work and research from his students, but the neuroscientist himself has no affiliation with the Australian startup.
