M. S. Aksoy
People give different answers when asked what ‘intelligence’ means. Some say it means ‘knowing a lot’; others say ‘thinking quickly’; or ‘putting things together in a particular way’. Psychologists also differ on the definition of intelligence. But it is agreed that among the important constituents of human intelligence are the use of sense, judgement, plausible, goal-directed reasoning, and appropriate knowledge and beliefs.
Two of the ways that people demonstrate their intelligence are communication and learning. Effective communication requires skills in the analysis of messages received and in the synthesis of messages transmitted. The ‘message’ may be a letter, an article, a poem, musical composition, painting, or indeed any other form of communication. In order to communicate effectively, one must be able to synthesize a message. To do that well needs the ability to make judgements about the level of sophistication of the recipient, careful use of the language of the communication and appropriate speed of presentation. Understanding a message also requires intelligence. A listener needs to know the meaning of most of the words being used and to have some knowledge of the context of the message.
The ability to learn and understand is considered by many as a vital component of intelligence and perhaps a definition of it.
‘Learning denotes the changes in a system that are adaptive in the sense that they enable the system to do the same task or tasks drawn from the same population more effectively the next time.’ ‘When a computer system improves its performance at a given task over time, without re-programming, it can be said to have learned something.’
There are many definitions for Artificial Intelligence or AI. One is: ‘the study of mental faculties that encompasses computational techniques for performing tasks which apparently require intelligence when performed by humans’. The term Artificial Intelligence was first used by John McCarthy in 1956. Since then, and especially in the last two decades, there has been a growing amount of research on AI. With many scientists, engineers and programmers either studying AI techniques or building AI systems or both, national and international organizations dedicated to AI have been formed and are growing. In the USA, for example, there is now an American Association for Artificial Intelligence. A number of AI applications in the medical and engineering fields have been successful and so established AI as a promising area of study with specialist sub-sectors, namely robotics, machine vision, natural language processing, machine learning, expert systems and neural networks.
Looked at closely, each of these sub-sectors turns out to be an attempt to imitate some human organs or faculties. Robotics, for example, aims to imitate what a human being can do using limbs, hands and feet, in co-ordination with the eye and brain-which is what ‘vision systems’ aims to copy. ‘Natural language processing’ is the development of systems that perform tasks using the kind of language that humans use in routine interactions. ‘Machine learning’ aims to make computers learn how to use computational techniques. An ‘expert system’ is built to house and sift large amounts of human knowledge in order to give advice in particular circumstances just as a human expert would do. And ‘neural networks’, as the name suggests, try to reproduce in part the human nervous system.
Even though there have been many successful applications devel
oped in these areas, AI, compared to human intelligence, is still in its infancy. Everyday observation shows that the modest brains of lower animals can perform tasks that are far beyond the range of even the largest and fastest modern electronic computers. Just imagine that any mosquito can fly around at great speed in unknown territory without bumping into objects blocking its path. And a frog’s tongue can catch these insects in full flight within a split second.
At present the number of processors which can be built for specialized parallel hardware is somewhere between 50,000 and 1 million. Compared to the number of neurones in the human brain this number is extremely small. The total number of neurones in the human central nervous system can only be estimated; some estimates put the number about 1011 combined with the average number of synapses per neurones this yields a total of about 1015 synaptic connections in the human brain, the majority of which are developed within a few months after birth.
Comparative facts like those show that there is a lot still to be learnt from human beings, animals and other creatures. When somebody looks at an intelligent system-say a vision system or an intelligent robot, they will remark that it is amazing. Indeed it is amazing to see a system correctly inspect and classify many products in the space of a second. Such systems are already installed and in operation on production lines and their performance is quite good. There are several expert systems in use in real world tasks. The importance of these systems derives from their artificial intelligence which in turn derives from human intelligence. A man-made system can be very smart and artificially very intelligent but no such system so far has been awarded a prize for its innovative abilities. It is the human being who made it who wins the prize. What is prized, what is of higher worth, is not the system but its maker or builder. What about the fantastic system that is a human being? Just as an AI system is strong evidence of the existence of its maker, namely, a human being, so too human beings and other natural systems are more and stronger evidence for the existence of their Maker. To make an artificial intelligence one must first have natural or real intelligence. That means the Maker of human beings and all other creatures has a supernatural power over all of them. Just as no one could say that AI systems build themselves, it should be impossible to say the same for human beings and other creatures. This supernatural power is the All-Mighty Creator. As we have seen, it is not the system but its maker that is prized and respected. Self-evidently, the One who created human intelligence is to be prized and respected more than anything or anyone else. God says in the Qur’an: ‘We have indeed created man in the best of moulds’ (95.3). As we believe that there is definitely a maker of an AI system, why should it be hard to believe that there is a Maker of human being?
- CHARNIAK, E. and McDERMOTT, D., (1985) Introduction to Artificial Intelligence, Addison-Wesley Publishing Company.
- HANCOX, P.J., MILLS, W.J. and REID, B. (1990)
- Artificial Intelligence/Expert Systems, Ergosyst Associates, London.
- MULLER, B. and REINHARDT, J. (1990) Neural Networks. An Introduction, Springer Verlag.
- FORSYTH, R. and RADA, R. (1986) Machine Learning:
- Applications in Expert Systems and Information Retrieval, Ellis Horwoood, Chichester.