Proving AI Is Possible or Impossible, Essay Example

Seemingly, present-day media and industries show notable success in applied AI and cognitive simulations; thus, one could argue that AI is guaranteed. However, strong artificial intelligence, duplicating human beings’ intellectual competencies, is highly debatable. Again, present-day challenges in scaling up or advancing AI’s modest accomplishments is another factor adding to debates of its possibility. Advanced AI is becoming an elusive aspect, influencing its chance even in a futuristic context. Research in symbolic artificial intelligence presently fails to offer credible proof that symbol systems can exhibit human-equivalent characteristics of general intelligence. Concurrently, connectionists’ analysis also fails to show evidence of AI models having a nervous system, even a simple one such as invertebrates.  Such pervasive deficiency of advancement in structuring a human-level AI adds to the challenging reality of establishing strong AI. Therefore, the question regarding the possibility of AI entails evaluating whether technological advancements can guarantee computers having a consciousness, an aspect that can be answered using the Chinese room contention.

It is still debatable on whether computers can think and act like humans. However, behaving like a human does not guarantee that a computer is a strong AI. Elements such as belief systems, hopes, fear, happiness, and pain are emotional and mental processes evident in humans. Thus, if a machine can qualify as having these mental aspects can be regarded as having a mind (Monin et al. 183). However, this assertion is still subject to debate because when assessing if an AI can possess a mind, one would have to deal with contentions that evaluate what makes up the mind. Accordingly, various philosophical models have emerged over the years to explain the fundamental nature of a mind. The functionalism philosophy has played a critical role in driving the notion that machines could possess a mind. For example, according to Piccinini “Functionalism plus computationalism equals computational functionalism. The same slogan is often understood to suggest, more modestly, that the mind is the computational organization of the brain—or that mental states are computational states—without the implication that such a computational organization is that of a program-controlled computer” (271). In line with this philosophy, intellectual states like belief systems are typified by functions or the objective it serves in an individual’s life. The aspect is otherwise regarded as its causal role. While varying functionalism forms exist, computational functionalism has had the most significant impact in debates regarding the AIs having a mind. Computational functionalism is linked to the mind’s computation theory. Under computational functionalism, a mental state is equivalent to a software state within computers (Piccinini 271). In this case, the brain symbolizes hardware and the mind to the software. Therefore, if the functionalism philosophy holds true, then one could argue that a definite possibility of computers having mental states. Unsurprisingly, most people contending in favor of the possibility of AI are grounded in the influences of the functionalism theory.

A similar influence towards the possibility of AIs has its roots in the belief that only fair intelligence measures can assess a mind’s likelihood. This reasonable criterion conviction inclines the need for a performance test. In this case, the underlying logic is that if a computer can exhibit intelligent behaviors such as those of human beings, it could be credited as having a human-level mind or behavior. If the computer successfully passes the intelligence test, it would be unfair to deny its human-level intelligence because it does not have a similar body organically structured. Such a proposition is supported by the notions against the appropriateness of intelligence discrimination. For example, history underscores discriminative practices. Some mainstream groups viewed other people as having inferior intelligence and not deserving any respect based on factors such as ancestries, race, or gender. Therefore, when judging intelligence, it would be prudent to avoid such mistakes with AIs if they prove that they have equivalent intelligence but are structured differently anatomically. In line with such an ideology, the Turing Test was developed to assess machine intelligence. Turing, the developer, admitted that even a machine regarded as intelligent based on the assessment might still fail to pass the mind test if it was unable to effectively imitate a human being (Warwick & Shah 990). Turing also identifies that the questions within the assessment were seemingly too vague. Therefore, if intelligent entities could fail the test, the evaluation could not function effectively as an intelligence definition. Again, the Turing test cannot successfully answer if machines have a mind because one can also develop a device with a fine-tuned array of canned responses (Warwick & Shah 990). On whether machines have mind to think the author provides a quote by Turing who says that “I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted” Turin 1950 (Warwick & Shah 990). Such machines would offer answers to the interrogator by referring to appropriate solutions ingrained in their system. Therefore, the Tt is not dependable when assessing the possibility of AI.

However, the evaluation of Chinese room contention by John Searle can help offer a nearly definitive answer to the likelihood of strong AI that has human-level capabilities. Searle’s Chinese room philosophy contradicts the functionalism model and does not support the Turing Test objective as a reliable assessment for machine intelligence (Monin et al. 180). In questioning the machine intelligence, the authors provide a quote by Searle “a computer is neither rational nor irrational, because its behavior is entirely determined by its program and the structure of its hardware” (Monin et al. 183). According to the writers, the Chinese room contention ascertains that programs cannot offer computers ‘a mind,’ comprehension, or consciousness irrespective of how human-like the programs make computers behave. The core of the Chinese room contention is a person visualizing themselves following a symbol processing program with English writings, otherwise regarded as the Turing test. The English-speaking individual who sits in the room aligns with English guidelines to manipulate Chinese symbols, whereas computers follow programs outlined in computing languages. The human being derives the aspect of comprehending Chinese by identifying the symbol manipulation guidelines; however, they do not, in that way, comprehend the language. In a similar vein, computers do what humans do, manipulating a symbol based on their syntax; thus, no machine can authentically comprehend Chinese by following programs. This Chinese Room contention is precisely directed at Strong AI, which is the stance that appropriately programmed machines or programs can understand natural languages and have other psychological competencies such as those of human beings whose behaviors they duplicate  (Monin et al. 183). In line with Strong AI, a machine can play chess with intelligence, establish crafty moves, or comprehend language. Monin et al. provides a quote by Searle against strong AI ““programs are formal (syntactic)” whereas “human minds have mental contents (semantics)” (180). Against such a reality, weak AI concerns a significantly modest contention that machines are just relevant in psychological and linguistic fields as they can stimulate mental capacities. The Chinese Room contention does not concern weak AI and neither uphold that technology can think. The philosophy contention is that symbols computation result in thoughts but do not necessarily prove that machines have a mind.

The Chinese room stances present the reality that no computing machine manipulating symbols grounded only in their formal syntactic elements such as shapes and positions can be regarded as genuinely comprehending language even if they succeed the Tt. Given that the Chinese room considers humans as biological machines understanding language without directly arguing that computers cannot comprehend language leaves room for the possibility of strong AI. However, the Chinese room philosophy insists that genuine thoughts and comprehension necessitate elements more than just computations, which present-day machines lack (Monin et al. 183). Again, scaling up today’s technologies or devices to attain such human-level competencies has proved challenging.

Given that the Chine room model does not reject any possibility of a machine having thoughts, then one can conclude that AI is possible if computation will be achieved together with human-level thinking, emotions, and mental processes. Presently, machines in configuration lack human-level understanding, and without such comprehension approaches, one cannot outline these machines as having thoughts. Without ideas, one cannot conclude that they can think. Lack of a mind in devices can lead to the argument that strong AI is yet to be achieved. While one cannot disregard the possibility of having machines with thinking and human-level mental state processes, it is logical to assume that it will take a long while.

The paper concludes that AI is possible but not in the future. Computation functionalism supports the possibility of AI, where a mental state is equivalent to a software state within computers. Also, the reasonable criterion argument inclines the need for a performance test. If a computer can exhibit intelligent behaviors such as those of human beings, it could be credited as having a human-level mind or behavior. However, the Chinese room notion rejects these philosophies arguing that programs cannot offer computers ‘a mind,’ comprehension, or consciousness irrespective of how human-like the programs make computers behave.

Works Cited

Monin, Maxim Aleksandrovich, and Vera Albertovna Terekhova. “The Chinese Room Argument in the Context of the Rational Action Theory.” Utopía Y Praxis Latinoamericana: Revista Internacional De Filosofía Iberoamericana Y Teoría Social 5 (2020): 179-186.

Piccinini, Gualtiero. “The mind as neural software? Understanding functionalism, computationalism, and computational functionalism.” Philosophy and Phenomenological Research 81.2 (2010): 269-311.

Warwick, Kevin, and Huma Shah. “Can Machines Think? A Report on Turing Test Experiments at the Royal Society.” Journal of Experimental & Theoretical Artificial Intelligence 28.6 (2016): 989-1007.