New Computer Chips That Think Like Humans

Ever since the first computer was created, not much has changed with how they process information. Sure they’ve gotten faster, smaller and way more functional, but if you move all that aside and just look at how a processor processes, you have the same exact thing: user inputs a function, function is stored in temporary memory, processor takes that function and performs a task.

The simple way to look at this process is to know that the processor and memory are still two very separate functions. Imagine if every time you wanted to do something, you had to write it down first and then go through the list one step at a time to make the task happen?! That’s what a computer is doing.

However, IBM has developed a computer chip where the memory and processing functions are one and the same. Darmendra Modha, the researcher leading the project, says, “It is IBM’s first cognitive computer core that brings together computation in the form of neurons, memory in the form of synapses and communication in the form of axons.” One example of how this will change the way we use computers is if you had a traffic signal being monitored by computers and video cameras. A brain-like computer can monitor these signals 24 hours a day and be able to detect anomalies (like traffic accidents) and be able to dispatch the needed support faster than waiting for an emergency call.

Another example would be to use such technologies in the oceans to monitor and detect abnormal changes in the temperature and overall condition of the water to know if things like rogue tsunamis are headed for a coastal town.

Researchers tell us that the brain-like chips are nowhere near the abilities of the human brain at this stage, but these new chips do show promise. Currently, they have 256 neurons-like nodes that allow the chip to contain 262,144 programmable synapses which is enough to drive a car through a simple maze.

My two cents

It’s kind of interesting and scary at the same time! On one hand, if a computer can make decisions for itself, we might be close to creating the perfect human, although it’ll be non-human if that makes sense. This could mean that it is always correct and nothing could corrupt it. However, it might be quite the opposite. If a computer can think, does that mean it can also be lied to? For example, if we got to the point where we created a robot with a real computer brain, could you tell it false information without it knowing whether it’s true? If so, could it really act on that mis-information just like a human does and make mistakes?

We might have a few years to wait before we know, but I can tell you we’re heading down a very interesting road in the technology sector.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.