What does it mean to be alive? This question has been haunting us since the beginning of time. Thousands if not millions of novelists, philosophers, scientists have tried to answer.
However, for practical purposes, you don't really need to know: you just live. You just learn to move in this world according to a certain set of rules, and as long as they work, you keep going.
All things considered this is not much different to the approach to brain-like computers that a Newark, California, based startup named Koniku is taking. Most of the experiments in this field are focused on trying to understand and replicate the infinite complexity of the brain using artificial methods, or on creating interfaces that connect the physical world with machines.
Koniku is taking another path, from the bottom up starting from the brain cells, the neurons, and trying to engineer them just like any other material to create chips that combine biological and artificial qualities.
The idea the is possible to code neurons is based on the fact that the way they communicate with each other using electricity spikes, is generally understood, although this is true only for certain kinds of neurons and certain classes of cells.
"Let's take a pyramidal neuron. We understand how spikes, the property with which neurons talk to each other, are initiated," Koniku's founder Oshiorenoya Agabi, a neuroscientist and bioengineer with several years of experience in this field tells me, "Do we understand this for generic pyramidal neuron? The answer is yes. The next question becomes, "How many kinds of pyramidal neurons are there?" Hundreds. Do we understand each of them? No. We don't."
But it doesn't really matter, Agabi believes, as long as it's possible to harness the power of biological neurons to do things that silicon chips currently can't. For instance, building drones that can 'smell" particles in parts per trillion, something that could be used to spot methane leaks in oil refineries (a first contract has been signed by Koniku with a customer interested precisely in this) or for other industrial, agricultural of military purposes.
"We are a business. We are not a science project," Agabi, who is scheduled to speak at the Pioneers Festival in Vienna, next week, says, "There are demands that silicon cannot offer today, that we can offer with our systems."
The core of the Koniku offer is the so-called neuron-shell, inside which the startup says it can control how neurons communicate with each other, combined with a patent-pending electrode which allows to read and write information inside the neurons. All this packed in a device as large as an iPad, which they hope to reduce to the size of a nickel by 2018.
"Traditionally, you will have a computer that has a sensor that is separate, and the computational engine that is separate. With our system, the sensor and the computational engine, and the control engine, they're merged into one," Koniku's founder explains.
For now, a first prototype of a "biological chip" able to store 64 neurons, enough to power simple applications, like chemical sensing, has been produced.
To create something more, a real computer made of biological cells, more time and money will be needed. As for the money, Koniku just raised an "undisclosed amount" that will help it focus on delivering the devices. As for the time, Agabi is rather bullish on this topic, saying he believes that Koniku is going be mainstream in two or three years and that they will be able to make a big breakthrough in the next five years.
It might sound overly ambitious, but Agabi is clearly deeply convinced of what he says, and relies partly on the fast-paced advancements of technology and partly on the creation of a community of computer scientists, neuroscientists, engineers and bio-technologists that will create product and applications on top of the Koniku neuron-based chips.
"At the moment, the wetware of which our brain is made is so advanced, that our software to drive it is not as advanced as the hardware. As academia and neuroscience progress, we update our software. We add more and more power to this wetware," he says.
For community building, the startup is inviting all interested developers to the Neurogrammer conference to be held on August 13-14th. Whether Agabi and his teams are real visionaries or just dreamers, only time will tell. But the fact that they have gauged the interest of big corporations like
