Human Neurons and A.I.

If the brain is a computer and the mind its workings, is this a fitting analogy of the computer and its software? What would happen if we had dedicated computers with a huge number of neuron circuits? Would intelligence develop? Would we be able to understand it?

Human neurons became the subject of interest of many AI scientists especially when most of the used AI methods were starting to fail. Some methods were proving to be inefficient and slow in evolution, while others were being proven wrong.
Scientists interested in creating a human-brain similar system started exploring more on the idea of using neurons as a prototype to follow and extract basic elements of its success. Some suggested the invention of something similar chemically while others wanted to start by developing simulators.

However, lately the trend became to extract real neurons and try to integrate them with computer chips creating a new type of computer hardware “neuro-chips” (Than, 2006). This experiment was repeated in several labs including the Georgia Institute of Technology and University of Padua in Italy. (BBC, 1999)
Even though these attempts are still very primitive and need much enhancements and further experimentation in order to develop, however, it seems most of the efforts of the young researchers are being dedicated for such discovery. Does this mean that if we create a dedicated computer with huge number of neuron circuits that this computer will be more developed and intelligent?

Though still a controversial issue and nothing is proven yet in this sense, however, many analysts are starting to believe more and more in this approach since they are finding similarities between the human bodies in general and computers. Example: “When we talk about psychotropic drugs, we use similar metaphors—the brain as computer, a neuron as a single switch, the brain as wetware containing software”(WOLPE, n.d.).

In neural networks, the number of neuron circuits tested upon only reached “several hundreds, compared to the 10^12 neurons that our brain possesses.” (Lecture notes) The results though are encouraging; still it is rather early to start judging the effect of huge neuron circuits.

Moreover, supposing such impact is possible, lots of questions arise: will the computer then be smarter than humans? Will we be able to understand its logic?
Scientists tend to support indications that could help prove the possibility of computers overcoming the human brainpower like in the case of the famous challenge in Chess between the different companies’ computers and Gary Kasparov, the famous chess master. Whether Kasparov lost once or twice only means that in this game the computer was able at least to challenge the best minds in the field. Lots of other examples exist.
As for the question about understanding the computer logic, I believe it will be quiet hard not to understand logic we have created/programmed; that is of course unless more chemical or human body elements are introduced.

I would like to end this by saying that this topic was maybe the hardest to tackle. Lots of controversies exist and nothing is proven yet. Supporting any idea while neglecting another opposing one will only get you to half the truth. Time will only prove which ideas will prevail and if really computers will ever have some sort of intelligence.


References:

1. Than Ker (27 March 2006). Brain Cells Fused with Computer Chip
[online] available from: http://www.livescience.com/health/060327_neuro_chips.html

2. BBC News (June 2, 1999) Sci/Tech Biological computer born [online] Available from: http://news.bbc.co.uk/1/hi/sci/tech/358822.stm
3. WOLPE PAUL ROOT (n.d.) Neurotechnology and Brain-Computer Interfaces- ETHICAL AND SOCIAL IMPLICATIONS [online] Available from:http://books.nap.edu/openbook.php?record_id=11083&page=58

Artificial Intelligence (A.I.)

When dynamite was discovered by Alfred Nobel (nobelprize.org, 2008), Nobel’s intentions seemed to be more directed towards helping mankind in minimizing the workload of time-consuming tasks. His main target markets were the drilling and construction industries. However, being such an asset for armies, dynamite soon became a rather deadly weapon that is still being used for killing hundreds of persons on a daily basis.

Reading the above makes one wonder about the several evolutions in the different science fields and how they affect our lives. In this regard, I would consider artificial intelligence (AI) one of the most dangerous and yet safe sciences.

AI is one of the controversial fields of modern time. Seeking to have another form of intelligence on earth that could someday trespass all its inventors’/discoverers’ own abilities is surely something to worry about. Knowing that computers have been replacing humans in many fields during the past decade makes one but wonder will they be able to replace humans as a whole? Will they reach a stage where they will feel that they can, and even need, to get rid of humans? "What we're trying to build are the mammals to compete with the big computational dinosaurs. You can imagine how the conversation went: 'They're too small. They're nothing - they're not enterprise scaled.' But the comet is coming. And when it does, we know who inherits the earth" (One Huge Computer). (Waldo, n.d.)

Let us consider the different possible dangers resulting from AI.
The first thing that comes to mind is the use of AI in producing new weapons of mass destruction. Same as with the dynamite example, advancements in technology were always reflected in the weapons’ industries – most of the time such advancements came as a result of new weapon creation funded researches.

Another effect of AI could be the result of unstable software or more specifically the presence of bugs. Let us consider the “possibility” that “smart bombs” (Harris, 2008) have certain bugs. This may lead to civilian casualties, mass destruction of cities even the death of the pilot firing the bomb himself.

What about ethics? Is it ethical to try to build a machine that could be more intelligent than such a divine race as Humans? Do we need such machine to exist? All are valid questions that are part of the big controversy surrounding AI.

However, considering the different aspects that AI has helped in, and the different solutions that were presented by its different approaches, surely everyone agrees that it is maybe the most beneficial field in terms of research results. Though results are not complete most of the times with respect to their initial aims, but lots of small discoveries on the way have made huge impacts in other different fields.
Consider the impacts of Fuzzy Logic on dishwashers; speech and image recognition systems on security checks’ systems; neural networks on air traffic and data mining systems…

Let me end this with a personal thought: I believe that it is in humans’ nature to search for new challenges and to try to understand or discover new things. Knowing this, research will continue, and especially in the two most controversial yet open-ended fields AI and medicine. Therefore, let us enjoy these evolutions and stop worrying about their consequences for these are inevitable.

Reference:

1. nobelprize.org (2008). Alfred Nobel - His Life and Work [online] available from: http://nobelprize.org/alfred_nobel/biographical/articles/life-work/index.html
2. Waldo Jim, Sun senior staff engineer, Java Developer (n.d.) The Dangers of Technological Progress: Potential Dangers –Tim Chao, Tuam Pham, Mikhail Seregine. Available from: http://cse.stanford.edu/class/cs201/projects-99-00/technology-dangers/future.html
3. Tom Harris. "How Smart Bombs Work". March 20, 2003 http://science.howstuffworks.com/smart-bomb.htm