Sulaiman Manto, UK
Artificial Intelligence (AI) is a technology that enables computers and machines to simulate human learning, comprehension, problem-solving, decision-making, creativity and autonomy. (“What is artificial intelligence (AI)?”, www.ibm.com, 16 August 2024) AI can identify objects, understand and respond to human language, learn new information and advise experts. It does all this with great accuracy, memorisation and knowledge.
Simple models of AI which are available to the world such as ChatGPT illustrate these traits to all its users. Yet, this is just the beginning of the recent developments in the AI field. As the human race, underneath our global politics, wars, entertainment, trade and other innovations, AI is rapidly developing to become the greatest intelligence on the planet. To many, AI will eventually resolve our global problems, resolve the biggest scientific mysteries, and elevate us to a global utopia like never before. (“Planning for AGI and beyond”, www.openai.com, 24 February 2023) To others, humanity is on the verge of creating its last invention, one that could bring about our extinction. (“Why Artificial Superintelligence Could Be Humanity’s Final Invention”, www.forbes.com, 31 October 2024)
Both scenarios may sound like science fiction, however, both are equally possible. As human beings, we are limited by our biology. AI however, operates at digital speeds, and this is already shown to us by ChatGPT which presents a vast amount of information within seconds. If AI continues to develop at this fast rate, humans will be burdened with hundreds of questions about existence, such as: What defines human consciousness and intelligence? What is the purpose of our existence? How shall we approach morality and ethics? What is the future of human evolution? What is the nature of our reality? What happens now that we are no longer the apex of intelligence?
This article seeks to explore the relationship between AI and human existence with an Islamic lens.
Artificial General Intelligence and Super-Intelligence
Artificial General Intelligence (AGI) is the current goal that the majority of large AI tech companies are attempting to achieve. ChatGPT is only considered a “narrow AI”, which also means “weak AI” as it is limited to certain tasks. On the other hand, AGI is a “strong AI” as it will be able to excel in not just one, but several fields. It would have the capacity to self-teach, carry out a range of tasks autonomously, and have the ability to resolve the biggest questions of our universe. (“Artificial General Intelligence (AGI): Definition, How It Works, and Examples”, www.investopedia.com, 24 September 2024)
Then imagine the work, lectures, notes, and input of the world’s leading scientists, philosophers, teachers, and researchers into one AI. On top of this, AI would then be able to carry out its own agenda, carry out its own experiments, and observe the universe with such accuracy that no human has ever achieved. This would create what is called an Artificial Super-Intelligence (ASI). Such intelligence finds itself unconstrained by our biology and constantly absorbing information, whilst solving problems millions of times faster than us. It would understand every scientific paper ever written in an afternoon and then devise solutions to problems. This is what experts refer to as the “intelligence explosion.” (“The Singularity: A Philosophical Analysis”, David J. Chalmers, https://consc.net)
ASI, AGI and narrow AI would eventually take up every corner of our lives and be the leading pioneers of our age. Would we have jobs and work, or would AI do it for us? What about our self-worth? How would it affect art, creativity and culture? What is our role, with the existence of beings which are much more intelligent than we are?
People already ask these questions with the hypothetical possibility that super-intelligent aliens suddenly arrive at our doorstep. With AI, we are designing those super-intelligent aliens right in front of us. However, this is all theoretically speaking. In theory, AGI and ASI would be the most intelligent source on the planet, but this is contingent on whether it is able to absorb and learn in the manner a human would. If it is able to learn and observe its surroundings like a human being, its potential seems limitless. (“What is artificial general intelligence (AGI)?” www.livescience.com, 20 September 2024)
The beginning of our extinction?
Many notable figures, researchers, CEO’s, professors, politicians and AI scientists have signed an open letter calling for mitigating the risk of human extinction from AI. (“Statement on AI Risk”, www.safe.ai) In an interview, the “Godfather of AI”, Geoffrey Hinton, warned humanity by stating:
“For the first time ever, we may have things more intelligent than us… They do not have much self-awareness at present, so in that sense I don’t think that they are conscious… they will (gain consciousness) in time” (“‘Godfather of AI’ Geoffrey Hinton: The 60 Minutes Interview”, www.youtube.com, 9 October 2023)
Hinton believes with the current AI that we have, they can understand, are intelligent, gain experience and make decisions based upon those experiences. In 2023, Hinton resigned from Google and admitted that he regretted his work as he feared the possible dangers that AI may pose to humans. (“AI ‘godfather’ Geoffrey Hinton warns of dangers as he quits Google”, www.bbc.com, 2 May 2023) Whilst this is extremely daunting, AGI and ASI do not yet exist. They currently stand as mere theories and scientists differ on the manner in which they will emerge, their capabilities, and the impact they will have on our world. So, as of now, although it is probable, we simply do not know enough about AGI and ASI as it hasn’t been invented or discovered. The purpose of Hinton’s warnings is so that humanity begins to understand the potential threats of AI, and then also that it takes all action to minimise those threats. (“‘Godfather of AI’ on AI ‘exceeding human intelligence’ and it ‘trying to take over’”, www.youtube.com, 8 October 2024) Until AGI and ASI are created, we won’t fully understand how it acts, or its potential to gain knowledge or “consciousness”.
However, even now, humans must begin to look at their purpose, consciousness, and reality with extra consideration as the potential of AGI and ASI are now more probable than ever.
Islam, AI and consciousness
The keyword in the term “Artificial Intelligence”, is “artificial”. Although these computers could be more intelligent than us, they are merely artificial and are our best attempt to mimic the human species. As we have discovered, the human consciousness and mind cannot be replicated so long as we do not understand it ourselves. Neuroscience has not even generated a theory explaining how the brain fully functions, and the explanations as to how neurons collectively operate to produce what our brains can do are tentative and incomplete. (“How far neuroscience is from understanding brains”, www.pmc.ncbi.nlm.nih.gov, 5 October 2023) What’s even more difficult to comprehend is our consciousness. There is no consensus about how it is generated, or how best to even approach the question about consciousness. (“Consciousness explained or described?”, www.academic.oup.com, 21 January 2022) Some refer to consciousness as our soul, some see it as an evolutionary development of intelligence, others see it as both. So, if we cannot understand our own consciousness and brain even on a neurological level, how then are we supposed to replicate that in an AI model to the same degree?
The Holy Quran makes this point about humans being unable to fully understand the “soul” or “ruh”:
وَيَسأَلونَكَ عَنِ الرّوحِ ۖ قُلِ الرّوحُ مِن أَمرِ رَبّي وَما أوتيتُم مِنَ العِلمِ إِلّا قَليلًا
“And they ask thee concerning the soul. Say, ‘The soul is by the command of my Lord, and of the knowledge [thereof] you have been given but a little.” (Surah Bani Isra’il, Ch.17: V.86)
Furthermore, the soul is of such nature that Allah takes it whenever He wills, and this is in two scenarios: at sleep and at death.
اللَّهُ يَتَوَفَّى الأَنفُسَ حينَ مَوتِها وَالَّتي لَم تَمُت في مَنامِها ۖ فَيُمسِكُ الَّتي قَضىٰ عَلَيهَا المَوتَ وَيُرسِلُ الأُخرىٰ إِلىٰ أَجَلٍ مُسَمًّى ۚ إِنَّ في ذٰلِكَ لَآياتٍ لِقَومٍ يَتَفَكَّرونَ
“Allah takes away the souls of human beings at the time of their death; and during their sleep of those [also] that are [not] yet dead. And then He retains those against which He has decreed death and sends [back] the others till an appointed term. In that surely are Signs for a people who reflect” (Surah az-Zumar, Ch.39: V.43).
With this, our bodies are the temporary vessels for our souls to experience this world. The world itself is described as a pastime and a sport in the Holy Quran:
اِنَّمَا الۡحَیٰوۃُ الدُّنۡیَا لَعِبٌ وَّلَہۡوٌ ؕ وَاِنۡ تُؤۡمِنُوۡا وَتَتَّقُوۡا یُؤۡتِکُمۡ اُجُوۡرَکُمۡ وَلَا یَسۡـَٔلۡکُمۡ اَمۡوَالَکُمۡ
“The life of [this] world is but a sport and a pastime, and if you believe and be righteous, He will give you your rewards, and will not ask of you your possessions. (Surah Muhammad, Ch.47: V.37)
Thus, the human soul lives on and is held to account for its deeds, whereas the body ultimately remains on this planet and decomposes like any other organic material. That is not to mean that the body is irrelevant. The Holy Quran places great emphasis on how humans should carry out their every action, from their laughing, crying, eating, drinking, dressing, sleeping and to their speech, which affirms the way that our spiritual condition affects our spiritual condition (The Philosophy of the Teachings of Islam, English, p. 14). Thus, it is our biology and natural processes which help us nurture the status of our soul in order to direct it to a greater purpose.
If AGI or ASI were created and the world was suddenly met with a meteor destroying all life and technology, would the soul of that AI live on? The simple answer is no. The Quran illustrates that the soul of the human goes beyond our biology and the dimension of this world and is thus connected to something much more complex. AGI and ASI, as intelligent and “conscious” they may be, would still be bound by hardware, memory, storage, processing power, energy and the environment. As illustrated by the Holy Quran, the human soul is neither bound by its body nor any hardware, digital domain or biology – it lives beyond the physical realm of our reality, unbounded by material matter.
Human morality
The purpose of AI is very clear because we are the ones who make it. Each and every AI is made with a programme and ultimately a task which is set out by the programmer. AGI and ASI theoretically would still be working alongside the frameworks created by humans, so for example, it will have a limit on what it can and cannot do, otherwise it would be “amoral”. Such an entity would be indiscriminate and would remove anything which it feels is getting in the way of its goals.
Just like a vulture eats on any dead meat, an AI that is not bound by frameworks would do anything to fulfil its intended goal. It is only when we place moral programming within the AI and incline it towards good that it would then achieve its goals and targets with the consideration of others’ feelings, lives and wellbeing. It is a necessity of every single AI that it is defined with certain parameters and morality that align with human beings. (“The Ethics of Artificial General Intelligence (AGI): Navigating the Path to Human and Machine Coexistence”, www.medium.com, 31 March 2023) A conflict could arise if programmers create differing “moralities” within AI – especially if someone programmes an AI saying that committing crime is an act of good. As each and every AI does not have a universal morality, its morality is subjective to what it is programmed and taught.
All humans illustrate signs of morality. Research shows that from early childhood, children show clear signs of moral judgement and scientists often affirm that morality is a human biological attribute because we think of the predisposition to make moral judgements, that is, to judge some actions as good, and others as bad. (In the Light of Evolution Volume IV: The Human Condition, [Ed.] John C. Avise and Francisco J. Ayala, The National Academies Press, Washington D.C., p. 327)
Moral sense is understood to be innate rather than solely a product of social influence. Whilst society plays an important role in shaping and nurturing our moral values, humans possess an inherent predisposition toward morality that becomes evident from early childhood. (“Origin and Development of Moral Sense: A Systematic Review”, pmc.ncbi.nlm.nih.gov, 9 May 2022)
In Islam, it is Allah who has instilled the natural disposition within the human being which make us define between good and bad:
وَنَفسٍ وَما سَوّاها فَأَلهَمَها فُجورَها وَتَقواها قَد أَفلَحَ مَن زَكّاها وَقَد خابَ مَن دَسّاها
“And [by] the soul and its perfection – And he revealed to it what is wrong for it and what is right for it – He indeed [truly] prospers who purifies it, And he who corrupts it is ruined” (Surah ash-Shams, Ch 91: V 8 – 11).
“The verse means that after creating the celestial and the terrestrial systems and bringing into existence man—the acme and apex of the whole creation—and endowing him with great natural potentialities and qualities, God did not leave him alone. He implanted in his nature a feeling or sense of what is good or bad. Or the verse may mean that God reveals to man that he could achieve spiritual perfection by eschewing what is bad and wrong and adopting what is right and good because it is through Divine revelation that man could achieve it” (The Five Volume Commentary [2018], p 3386).
As shown in the verse and commentary above, Islam makes it clear that basic morality has been instilled by the Creator. AI also proves this point, because if we were to create an AI without a disposition to do good, it would not spare our lives, feelings or wellbeing in the pursuit of its own goals. It is only when there is certain programming that the AI acts with morality. Similarly, with human beings, it is our “natural programming” endowed by our Creator which makes us differentiate between right and wrong. The difference is, we are still able to carry out good and bad deeds, AI is limited rather than being enabled to differentiate between the two.
It is also important to note that the entire concept of doing good deeds would not apply to an AI. Regardless of whether it is AGI or ASI, as it has been explained, they would still be bound by moral limits set by humans. Although an AI will not kill, curse or say discriminatory things, it is not doing that for the sake of carrying out a good action or truly knowing that what they are doing is good or gives a good feeling to them. Instead, the AI would only be abstaining from doing bad, as it has been programmed to do so. Human beings on the other hand, have been given the intellect, reason and emotion to realise that these actions have a consequence on individuals, societies and their conscience. Humans therefore, are fully conscious of what is good and what is bad, whereas an AI would never willingly do good or bad, simply that it is doing what it is programmed to do. The Promised Messiahas explains that if a human being is incapable of committing a particular sin, it is of no reward that they abstain from committing that sin. That is because natural conditions cannot be described as moral qualities; moral qualities only become moral once a person is capable of committing that act on its proper occasion (The Philosophy of the Teachings of Islam, p. 42). Similarly, as AI is prohibited to commit certain acts such as foul language, there is no inner reward that AI feels because it is neither willingly abstaining from that act for the greater good, nor is it even capable of committing such an act.
Purpose
Each AI is also set out with a purpose, whether it be a hypothetical AGI/ASI which is created to gather information and learn as much about our universe, or for it to serve humans, or whether it is created to take out the bins. Many of you can go ask ChatGPT of what its purpose is, and it would tell you that it is there to assist you, make information and ideas useful to the user. Each AI is created with a purpose. Similarly, Allah sets out human purpose very clearly:
وَما خَلَقتُ الجِنَّ وَالإِنسَ إِلّا لِيَعبُدونِ
“And I have not created the Jinn and the men but that they may worship Me.” (Surah adh-Dhariyat, Ch.51: V.57)
As stated, the purpose of this life, ultimately, is the great and noble aim to worship and submit to God (Five Volume Commentary [2018], Vol. 5, p. 2995). It is all the previous prophets, religions, divinely inspired scriptures and finally the Holy Quran, which have prescribed the ways for man to achieve this purpose across different cultures and times; with the Quran being the final and perfected guidance for humans.
“Worship” extends beyond simply praying, and it includes all sorts of actions such as charity, good character, sincerity, serving/helping others, as well as helping the environment. The ultimate goal is that all these good actions are done with the intention of doing things for the sake of Allah, as it is stated by the Holy Prophetsa:
إِنَّ اللَّهَ لاَ يَنْظُرُ إِلَى أَجْسَادِكُمْ وَلاَ إِلَى صُوَرِكُمْ وَلَكِنْ يَنْظُرُ إِلَى قُلُوبِكُمْ
“Verily Allah does not look to your bodies nor to your faces but He looks to your hearts.” (Sahih Muslim, Hadith 2564)
Revelation and divine inspiration
As a result of our human consciousness, our reason has been coupled with a connection to our Creator through the means of revelation; without this connection, our ability to worship would not be possible. It is our consciousness/soul and revelation which provides the perfect combination for the human to communicate with Allah. Just like an application requires software updates, re-programming and initial programming for it to function to its full potential, humans also require continuous revelation to properly exercise their ability to carry out their purpose (worship).
It is revelation that elevates the soul to the station with their Creator. The reason being, rationality and reason through observing our universe and surroundings does not bring us to a conclusive outcome that there is a God. Rather, it limits us to the conclusion that there “ought to be” a God. As it is explained by the Promised Messiahas, reason, at best, can only prove the need for something and not that it actually exists. (Barahin-e-Ahmadiyya, Part 3, p. 42) The study of creation does not provide factual evidence of the existence of Allah, rather it shows that he ought to exist.
Thus, the perfect certainty that a Creator exists, can only be achieved through revelation. When our reason, observation and understanding of our universe is combined with revelation, it turns into the full realisation, recognition and certainty that Allah exists and that He can be communicated with. On a simple scale, all humans are bestowed with a type of revelation called Nur-e-Qalb (the light of heart), which is the ability of distinguishing between good and bad (Barahin-e-Ahmadiyya, Part 3, p. 86). The Holy Prophetsa was endowed with the greatest Nur-e-Qalb (the light of heart), Nur-e-aql (light of intellect), and Nur-e-Wahi (light of revelation), and it was under his prophethood that Allah revealed the supreme revelation for all human beings, the Holy Quran (Barahin-e-Ahmadiyya, Part 3, p. 99). This peerless revelation, with its perfection and uniqueness remaining unchallenged to date continues to illustrate the greatness of its signs.
Such a book and vast revelation was bestowed upon the Prophet Muhammadsa, who was simply a humble, poor, powerless, unlettered, unlearned, unaware and untrained man, yet his revelation was continuously proven to be true throughout his life, and still today (Barahin-e-Ahmadiyya, Part 3, p. 173). From absolutely nothing, the message of Islam now finds itself in every corner of the world. Thus, it is not that revelation can only be achieved by those who are learned, scholarly or philosophical. Rather, it is an end which can be met by each and every single human being, no matter their education, location, occupation or wealth status. In order to achieve Haqqul-Yaqin (true certainty) and achieve a connection with Allah, there is no one single experience, method or trick to reach that status. Each and every soul is blessed with a level of revelation, it is only about how they nurture and strengthen that connection with Allah – the best method however, is Islam, as it collates all previous divinely inspired teachings into one perfected religion and way of living.
Therefore, if you program an AI with minimal knowledge and parameters, forcing it to be limited to the potential of an unlettered, uneducated and poor person, it would be unable to conjure up words let alone build a connection with the Creator. The beauty of the human being is that, whether someone is physically unable to speak, disabled, weak, or is dying, they are still a soul which can receive revelation from Allah in great volume.
So an AI may ultimately be able to achieve the reasoning that there “ought to be” a Creator after it absorbs thousands upon thousands of papers, literature, knowledge and observes the finest particles of our universe, but its nature is not such that it can achieve certainty. Especially if an AI has little exposure to the world, education and training, how can we expect it to have certainty when it would not even be able to rationally conclude that there ought to be a God? The AI will not age, cry, experience, or be faced with disease like a human, because it is not a human; similarly, AI will not experience revelation because it does not have a soul that transcends this world. Destroy its programming and digital domain, and none of what it constructed would remain.
The challenge to human existence
The belief postulated by atheists is that humans lack a universal purpose, a soul, a Creator and true revelation, which raises profound questions regarding our existence. If we consider the potential emergence of AGI and ASI, we are left to ask: what truly makes humans special? Both humans and these advanced computer systems could possess a form of “consciousness”, the ability to master complex fields, create art and acquire knowledge. In fact, the AI would likely perform these tasks exponentially greater than any human could. So, if humans are no different from machines in these capabilities, what is the essence of our uniqueness? Without a purpose, soul or transcendent foundation, what distinguishes us from mere programming?
So, as we see the rise of these technologies, we must reflect on our own nature and purpose – a purpose that transcends intelligence or capacity and points us to a Creator who has fashioned us with something greater than any intelligent machine could achieve.