Artificial intelligence (AI) sometimes invokes imagery of the distant future where robots live among humans, or evokes emotions of fear of machines taking over society. Even Stephen Hawking and Elon Musk have warned of AI’s threats.

AI is the simulation of human intelligence processes by machines or computer systems. Some of the activities computers with artificial intelligence are designed for include learning (acquisition of information, and the rules for using the information), reasoning (using rules to reach approximate or definite conclusions), self-correction, speech recognition, planning and problem solving. The bulk of the AI development today uses human reasoning as a guide to provide better services or create better products rather trying to achieve a perfect replica of the human mind (link).

Understanding AI further

There are several ways of classifying AI. Two broad main classifications (link):

1. Weak AI or Narrow AI:

It is an AI system that is designed and trained for a task. An example would be a poker game where a machine beats human where all rules and moves are fed into the machine. Here, every possible scenario needs to be entered beforehand manually. Virtual personal assistants, such as Apple’s Siri, are a form of weak AI. Every weak AI will contribute to the building of strong AI.

2. Strong AI:

Also known as artificial general intelligence, is an AI system with generalized human cognitive abilities. When presented with an unfamiliar task, a strong AI system can find a solution without human intervention.

In addition, AI can also be categorised into 4 functionalities types (link):

• Type 1: Reactive machines
  An example is Deep Blue, the IBM chess program that beat Garry Kasparov in the 1990s. Deep Blue can identify pieces on the chess board and make predictions, but it has no memory and cannot use past experiences to inform future ones. It analyzes possible moves — its own and its opponent — and chooses the most strategic move. Deep Blue and Google’s AlphaGO were designed for narrow purposes and cannot easily be applied to another situation.

• Type 2: Limited memory
  These AI systems can use past experiences to inform future decisions. Some of the decision-making functions in self-driving cars are designed this way. Observations inform actions happening in the not-so-distant future, such as a car changing lanes. These observations are not stored permanently.

• Type 3: Theory of mind
  This psychology term refers to the understanding that others have their own beliefs, desires and intentions that impact the decisions they make. This type of AI should be able to understand people’s emotion, belief, thoughts, expectations and be able to interact socially. Even though there are a lot of improvements in this field, this kind of AI does not yet exist.

• Type 4:Self-awareness
  In this category, AI systems have a sense of self and consciousness. Machines with self-awareness understand their current state and can use the information to infer what others are feeling. This type of AI does not yet exist.

Problems with AI

The core problems of AI at this point in time include programming computers for certain traits such as knowledge, reasoning, problem solving, perception, learning, planning, and the ability to manipulate and move objects (link). While AI tools present a range of new functionalities for businesses, the use of AI also raises ethical questions.

Selection of data – This is because deep learning algorithms, which underpin many of the most advanced AI tools, are only as smart as the data they are given in training. Since a human selects the data to be used for training an AI program, the potential for human bias is inherent and must be monitored closely.

Amount of data required – Machines can act and react like humans only if they have abundant and sufficient information relating to the world. AI must therefore have access to objects, categories, properties and relations between all of them to implement knowledge engineering. Initiating common sense, reasoning and problem-solving power in machines is a difficult and tedious task.

Proper classification of data – Learning without any kind of supervision requires an ability to identify patterns in streams of inputs, whereas learning with adequate supervision involves classification and numerical regressions. Classification determines the category an object belongs to and regression deals with obtaining a set of numerical input or output examples, thereby discovering functions enabling the generation of suitable outputs from respective inputs.