In a move to improve the learning capabilities of computers that we use today, Qualcomm has announced the Zeroth NPU. They state that the goal of this NPU is to create a new computer processor that mimics the human brain and nervous system so devices can have embedded cognition driven by brain inspired computing. I wonder why they picked the Zeroth name when you consider that Isaac Asimov had a Zeroth law in his Three Laws of Robotics. For those unfamiliar, the 0th law or Zeroth law, is "A robot may not harm humanity, or, by inaction, allow humanity to come to harm."
The name that Qualcomm used is a little ominous if you ask me, but I guess it also proves the point that Qualcomm is trying to make with the Zeroth. They want to help our devices become more learning capable and to see the world and things in the same ways that we do in order to deliver more compatible experiences. Perhaps we aren’t building machines to think entirely like humans, but we are hopefully at least showing them how we see the world in order for them to think beyond binary functions.
The video above is a little interesting because it shows how they are able to help a robot to learn via positive feedback (like a living thing would learn). However, the whole Good Robot things is a little scary because what happens if it’s a ‘Bad Robot’. While I really like the fact that Qualcomm is getting involved in things like the FIRST Robotics challenge and building an NPU for more compatible robots, I will always worry about how smart these robots can get if we give them enough cognitive capability and processing power.
The creation of a Neural Processing Unit is a lofty one and will most likely not see any results for many years to come, but it is certainly interesting to see Qualcomm trying to bridge the gap between humans and the processors in their hands. Surely, having the ability replicate the human nervous system and some of their thought processes is a good thing to have, but I just hope that there are some very strict checks and balances within these systems. You know, to prevent a Skynet-like event where the robots become self aware and start to realize that the world is a better place without us. It’s a crazy thought, sure, but giving computers the ability to think and feel like humans is also a bit crazy too.