Disadvantages of Artificial Intelligence

Artificial intelligence systems are based on the ability of a machine to learn. Babies for example are born with the ability to learn and seek out knowledge, in the same way if an artificial intelligence system is created, it too will be the equivalent to a new born baby in the sense that is will learn and seek out knowledge. Although the human brain's ability far surpasses any computer, if AI was created and it was powerful enough it would be just like a baby. It could be taught to be good or bad. A great example is used on the following website, http://tech.suramya.com/about_AI/, in it the author states, "Because AI is learning based, there is a fearful thought in mind, will machines learn that being rich and successful is an advantage, then rage war against economic powers and then control the world? This is not unreasonable but as every major religion and philosophy tells us, you cannot appreciate happiness without experiencing sadness. If we are going to create, or help create, a truly intelligent being, then we have to accept that it will learn to hate, as well as to love. It will know bad, as well as Good, Wrong, as well as Right". Because many theorists and scientists believe that it is not unreasonable that AI can go wrong there have been propositions to develop AI with safeguards so that we can shut them down in case they turn against humans or do not obey commands. In recent years there has also been the development of a non for profit organization named "The Singularity Institution for Artificial Intelligence", they can be found at http://www.singinst.org/.

Sources

http://tech.suramya.com/about_AI/
http://www.singinst.org/
http://www.buzzle.com/articles/disadvantages-of-artificial-intelligence.html

Advantages of Articial Intelligence

There exist countless advantages to developing advanced AI technologies. A few examples are:

• Control (Air Traffic)
• Debugging
• Design (Computer Configuration)
• Medical Diagnoses
• Instruction/Training
• Interpretation (Speech)
• Monitoring (Nuclear Plants)
• Planning (Mission Planning)
• Factory Scheduling
• Prediction (Weather)
• Repair (Telephone)
  

(The image on the left hand side shows advantages of using an AI expert system as opposed to a human expert.)

As you can see, there exists many applications and advantages to using AI technology. For example medical diagnoses AI technologies can really help cut down wait times at hospitals and clinics. An expert system is defined as "computer software that attempts to mimic the reasoning of a human specialist" (Retrieved from http://www.merriam-webster.com/dictionary/expert+system, on December 1st, 2008. ) By implementing AI expert systems patients who are only experiencing minor discomfort can be diagnosed by AI expert systems and not have to wait for a doctor, thus reducing waiting times in hospitals and clinics because of less patients.

AI can help industries in other areas as well. Speech recognition is one of them. Right now speech recognition works with computers being able to match an input sound with the words associated with the sounds in their databases. The drawback here is that not everyones voice is the same and people have different ways of talking therefore it is not a fool proof method of speech recognition. However, if computers can learn through AI what the context of the word is and what the topic is and analyze the speech patterns of a person and more, they will be able to fill in words that perhaps they do not understand with the most logical word to fit in the sentence. Right now Microsoft word does a little bit of this but nothing too advanced. If a computer is taught to learn and use "judgment" speech recognition softwares will become very accurate and reliable. This can help industries because it speech recognition can help speed up data searches and computer use. Imagine being able to tell your computer that you had a project to do on AI and the computer looks up all relevant information and presents it in a clear and concise document for you to read without having to search for it yourself. That is the future of AI!

Sources

http://www-formal.stanford.edu/jmc/whatisai/node3.html
http://library.thinkquest.org/2705/
http://www.cse.msstate.edu/~hansen/classes/AIspring04/slides/expertsystem.pdf
http://www.washingtonpost.com/wp-dyn/content/discussion/2006/06/14/DI2006061402086.html

Future of Artificial Intelligence

The future through the eyes of artificial intelligence can be seen through many perspectives. Artificial intelligence is brought through the development of many different sectors in an attempt to enhance the way of life for many including the financial departments. Ray Kurzweil, an author focusing on AI, Dr Rodney Brooks, a director of MIT's artificial intelligence laboratory and Colin Angle, CEO of the company iRobot were interviewed on a special BBC News Online by reporter Kevin Anderson. Kevin Anderson reports that Dr Brooks believes “that artificial intelligence is about at the same place the personal computer industry was in 1978” [Brooks, 2001]. The researchers also acknowledged that the future of artificial intelligence is but a very slow process and will take time before it become exclusively significant.

Artificial Intelligence rather than replacing current technologies will in the future enhance and evolve current technologies. For example, currently in car's computer chips control certain function. AI technology has the potential to take over ALL functions of a car and can eliminate the need for a driver as well. This can benefit society greatly if we are successful because AI technologies can become more efficient and safer drivers if taught to do so. Therefore, AI technology can be used to evolve current technologies and make them more efficient. AI technology can also be used to replace certain jobs currently held by humans. For example, currently weather forecasts are created by gathering environmental data and having a meteorologist analyze it and come up with certain conclusions. On the other hand if AI technology becomes integrated as part of the weather forecast process, the technologies may be able to more efficiently and more accurately process the tons of information gathered through the computer systems and may make weather forecasting much more accurate. Thus, this would effectively eliminate the need for humans meteorologists.

Sources

http://news.bbc.co.uk/2/hi/in_depth/sci_tech/2001/artificial_intelligence/1555742.stm
http://www.computing.dcu.ie/~humphrys/newsci.html
http://www.scienceclarified.com/scitech/Artificial-Intelligence/The-Future-of-AI.html

History of Artificial Intelligence

The topic I have chosen to blog about is Artificial Intelligence. According to Webster's Dictionary, Artificial Intelligence is "the capability of a machine to imitate intelligent human behavior". In other words it is the effort of the science community to create a machine with the ability to think and act like an intelligent human being, and most importantly learn. Currently "Researchers are creating systems which can mimic human thought, understand speech, beat the best human chessplayer, and countless other feats never before possible" (http://library.thinkquest.org/2705/, Retrived on November 28th, 2008). Since human beings use their brains to steer their behavior and thought the greatest challenge facing the scientific community is designing a robot system that mimics the human brain, which is made up of billions of neurons interconnecting to produce thought. This is one of the greatest challenges of attempting to create artificial intelligence during our time.

Artificial technology however has come a very long way since it's beginnings. A complete time line from start to present is available on USA Today's Tech section. The notion of AI began in the 1950's when Alan Turing published his book "Computing Machinery and Intelligence". Later on, in 1997, Deep Blue, IBM's AI computer project, beat beats world champion Garry Kasparov in chess match. This was followed by the first commercial interactive robot using AI technology to become commercially available to the public in 2000.

There exists a very big difference in the state that AI research is at now and the idea of robot's like R2D2 from Star wars. The technology needed to power thought and advanced learning in computers and/or robots is non existent as of now. According to particle physicist Michio Kaku the most advanced robotic artificial intelligence system that exists today is the equivalent of a "retarded cockroach". Meaning, the AI ability of machines today is not very advanced. If you give them a set of instructions, they will follow. For example, the robot that is currently on the lunar surface collecting samples and data, according to Michio Kaku, "...would take an hour to travel from one side of the room to the other side". Therefore, research into AI does have a long way to go but considering the strides in computer systems development that the world has gone through in the past decade, it is possible that AI may be properly developed in the not so distant future.

Sources

http://library.thinkquest.org/2705/
http://www.livescience.com/artificialintelligence/
http://www.aaai.org/AITopics/pmwiki/pmwiki.php/AITopics/BriefHistory
http://www.usatoday.com/tech/news/2001-06-20-ai-history.htm