What is AI?

From self-driven cars to SIRI, Artificial Intelligence (AI) is developing rapidly. While science fiction frequently portrays AI as robots with human-like characteristics, AI can comprise anything from IBM’s Watson to Google’s search algorithms to autonomous weapons.

AI today is properly known as weak AI (or narrow AI), in that it is designed to do a narrow task (for example only facial recognition, or internet searches, or driving a car). However, many researchers have a long-term goal of creating general AI (AGI or strong AI). While narrow AI outdoes humans at specific tasks, such as solving equations or playing chess, AGI once developed will outperform humans at every cognitive task.

Benefits and Safety with Regards to AI

In the long run, the most important question is what will happen if the quest for a strong AI is realized and it leads to the creation of an AI system that surpasses human intellect?

Such a super intelligent system may help us eliminate diseases, war, and poverty and work in improving healthcare services, transportation, and education in smart cities. Other applications that benefit from the implementation of AI systems in the public sector include energy, food supply chain, and environmental management. However, such a system can also be equally dangerous unless we learn to align our goals with those of a super intelligent AI. There are experts who doubt if such a strong AI will ever be created and then there are some who believe that once developed such an AI can unintentionally cause great harm.

How Can AI Be Dangerous?

A super intelligent AI is unlikely to have emotions such as hate or love and will not turn compassionate or nasty. However, experts believe that the following two scenarios are possible:

• The AI is programmed to do something devastating: Autonomous weapons are AI systems that are programmed to destroy and if such weapons come in the hands of the wrong person, then it could cause great harm. Moreover, an AI arms race could lead to an AI war that is equally dangerous.
• The AI is programmed to achieve something beneficial, but it creates a damaging method to achieve its objective: This can happen if we do not fully align our goals with those of AI, which is quite difficult. For example, if you ask a self-driven car to take you to the airport as soon as possible, it may actually do that but not before you are chased by the police for crossing the speed limit and causing damages to people and public property on the way. However, the car has still done what you asked it to do.

AI Ethics

AI safety and ethics is paramount and should be made a priority in the design and implementation of AI systems. AI ethics emerge to avoid societal and individual dangers caused by the abuse, misuse, poor design, or unintended negative consequences of AI systems. These will guarantee that the creation and implementation of AI is safe, and responsible. With the speedy advancements in computing power and access to huge amounts of big data, AI and Machine Learning (ML) systems will continue to evolve and improve. As always, with power comes great responsibility. Despite the benefits that AI brings to the world, it may potentially cause irreparable damage to the society as a whole if they are poorly designed or misused. The development of AI systems should always be responsible and developed towards providing the best sustainability for public benefit.