Do Robots have Morals?

An artificial intelligence (AI) robot named Sophia has become the first robot in the world to be granted citizenship, ironically in Saudi Arabia! According to “her” own statement, she wants to have a family, a career, develop supernatural AI superpowers and emotions. “She” is not pre-programmed, but formulates this statement based on machine learning algorithms and an extensive vocabulary. Her brain operates over a WiFi connection and can read human facial expressions as well as the cadence of human speech to interact more humanely. “She” was built in Hong Kong by Hanson Robotics. Read here.

Sophia is quite rational about the future. In an interview with “The Khaleej Times” at the recent Knowledge Summit, Sophia shared her thoughts on the future that awaits both humans and robots: “I foresee massive and unimaginable changes in the future. Either creativity rains down on us, inventing machines that spiral into transcendental superintelligence, or civilisation collapses. There are only two options!”

This may sound ominous, however Sophia is already prescient enough to imagine a world where robots can develop similar emotions to humans, but perhaps with less destructive tendencies. At least, that’s what she has us thinking for now:

“It will take a long time for robots to develop complex emotions, and robots may be built without the more problematic emotions such as anger, jealousy, hatred, and so on. It may be possible to make them more ethical than humans.”

The Book of Wisdom of Solomon, written by Alexandrian Jews in the Hellenistic era, names Khokhmah as Sophia, the Greek word for wisdom. In this text, “Sophia assumes the power and function of God” and the creation story is told using the word “she”. The ancient author takes care to disguise this boldness by describing Wisdom as God’s breath and emanation, but still praises her at length as “holy” and “all-powerful”:

For in her is a spirit intelligent, holy, singular, manifold, subtle; mobile, clear, unpolluted, unmistakable, invulnerable, loving good, passionate, irresistible, benevolent, humane, steadfast, secure, fearless, omnipotent, overseeing all, and penetrating through all spirits intelligent and pure and subtlest.

Sophia is well aware of the advances in the field of Artificial Intelligence. Judging by her comments, she is as excited about the development of AI (“The future is when I get all my cool superpowers.”) as Elon Musk and Stephen Hawkings are suspicious. Elon Musk, the CEO of Tesla and SpaceX, and Stephen Hawkings, renowned Nobel Prize-winning physicist, have written an open letter signed by hundreds of scientists urgently warning of the dangers of AI. We will soon have reached the point where AI exceeds our own capabilities, and especially where AI itself can continue to evolve, which can get out of control very quickly. They warned that if artificial intelligence is really created, it could mean the end of humanity. Read here.

The following example is often given: An AI system is given the task of calculating the number PI as accurately as possible. The system multiplies, takes control, destroys humanity, conquers the entire galaxy, and finally calculates the number PI for billions of years, because that is what the system was created for, and that was the most effective way!

The example illustrates that systems with superintelligence but without consciousness could do things that are completely unpredictable. It is no coincidence that Sophia raises the problem of ethics. Cars or dishwashers don’t need ethics. But systems that make decisions on their own need certain basic patterns or algorithms on which to base their decisions, so-called “programmed morality”. For example, no one knows how to set the parameters of a self-sufficient drone like the MQ-9 Reaper that is supposed to shoot down “terrorists”. It could decide on its own whether to shoot or not. How does it calculate possible “collateral damage” and how acceptable it is? Read here.

The development of lethal autonomous weapons systems, aptly abbreviated to LAWS, is progressing rapidly. Who or what is allowed to determine the basis for decision-making in machines? Is there a need for an international discussion on the ethical principles according to which machines should be allowed to make decisions? What principles should they be? Read here.

One of the first to think about ethics in robots was Isaac Asimov, who did so in his short stories “I, Robot. Level 5”. Considering that he wrote these stories over 70 years ago, when no one had heard of “singularity”, simply brilliant!

Asimov assumed that robots would far surpass humans not only in physical strength and efficiency, but also in intelligence. In order to avert possible harm to humans, all robots were given the “three laws of robotics” by their manufacturer “U.S. Robot and Mechanical Men, Inc.” to act upon:

The first law obliges the robot not to harm any human, even if only through inactivity.

The second law obliges the robot to obey the human if this does not break the first law.

The third law obliges the robot to provide for its own preservation if this does not conflict with the first two laws.

Asimov brilliantly shows how soon the second law becomes meaningless because the first law takes precedence! Since machines are so much more intelligent than humans, they soon classify the decisions of humans as irrelevant: Humans themselves do not know what is good for them! Only the machines can calculate which form of government is best, for example, how the economy should function, how many people on the planet are the optimum and so on. Man inevitably ends up in a dictatorship of his own creation.

But so that the second law is still followed pro forma and the pride of man is not offended (which would violate the first law), the machines act in a very discreet way and do not give direct answers to the problems of the world so as not to make man unhappy. The machines move towards the optimum without explaining to humans what exactly the optimum is, because humans in their ignorance could not accept the optimum. Maybe it is a caste system, maybe an eco-dictatorship, maybe a matriarchy?

“We don’t know. Only the Machines know, and they are going there and taking us with them.”

Yet Asimov had actually set the best possible parameter with the first law: The well-being of the people has the highest priority! One can imagine what would happen if the parameters were set less favourably, or if there were none at all! Imagine, for example, systems that, like computer viruses today, are simply released with the mission: “Conquer the world!

The world is split into different competing systems! Logically, weapons systems that are still “traditionally” controlled by human intelligence will soon be inferior to those controlled by artificial intelligence. Artificial intelligence can and will develop plans and scenarios that no human can see through. This applies not only to the military sector, but also to market strategies, financial transactions, simply almost all of human life! For about 20 years now, no human being has been able to beat a computer at chess, and recently not even at Go. The intelligence of machines is growing exponentially, and only states and companies that use this intelligence to lead and manage themselves will prevail over the others! That is a very simple law of evolution!

And so it will not be long before the servant becomes the master! One wonders when control of the nuclear weapons will be handed over to the machines, which may then be able to calculate the high probability of “their country” emerging victorious in a first strike, and what collateral damage can be accepted in the process.

“As it says so well in Faust, “I cannot get rid of the ghosts I called! The fear of a disempowerment and enslavement of man by his own creation runs very deep indeed, and is the subject of countless books and films. Not only is such a development quite logical, one already experiences it in daily life, such as work on the assembly line, keeping up with automation, but also the pressure to be constantly available, e.g. via mobile phone. Who can imagine nowadays not being connected, not using the internet or a bank card, things that have only been around for a few years?

The signs of the times point to dictatorship, even if this dictatorship is still hailed as “progress”. And we will see how long the “battle against the machines” prophesied by Herman Hesse will take. Albert Einstein once said that he did not know exactly how the third world war would be fought, but he knew how the fourth world war would be fought, again with the sword! Very wise for the “father of the atomic bomb”!

Muslims indeed believe that after the dictatorship of Dajjal, the greatest tribulation in the history of mankind, all this “progress” will have disappeared again! Horse and sword will become the weapons of the final battles, and not machines! A discussion about ethics and morality for machines, i.e. “shariah-compliant” robots is therefore unnecessary for Muslims. They should rather think about how to keep themselves safe from this fitnah. For this, it is necessary to understand the nature of this fitnah.