Weak vs. strong AI

Weak vs. strong AI – a definition

What distinguishes strong AI from weak AI? What are the characteristics and how is the dividing line defined?

Intelligence is generally characterized by targeted perception of the environment, rational reasoning (including autonomous learning and problem solving) as well as purposeful acting (including communication by and between intelligence media).

Artificial intelligence

The term artificial intelligence implies the creation of a machine-based replication of intelligence oriented towards the intelligent capabilities of humans. In such a machine-based replication, there is a typical division into two categories: weak AI and strong AI.

Weak AI

Weak AI (also known as narrow AI) does not exhibit any creativity, nor does it have the explicit ability to independently learn in the universal sense. Its learning abilities are mostly limited to training of detection patterns (machine learning) or comparison and search operations with large quantities of data.

Using weak AI, clearly defined tasks can be handled based on a defined methodology in order to solve more complex problems which, however, are recurrent and precisely specified. The benefits of weak AI are especially relevant in automation and controlling of processes as well as in speech recognition and processing. For example: Text and image recognition, speech recognition, translation of text, navigation systems, etc.

Digital assistant systems like Alexa, Siri and Google Assistant also belong to the category of weak AI.

Strong AI

Realization of strong AI is not yet within reach: The objective underlying the concept of strong AI is to allow natural and artificial intelligence media (e.g. humans and robots) to establish a level of mutual understanding and trust when working in a joint field of activity.

Thus, efficient human-machine collaboration could be learned and facilitated, for example. Strong AI is able to independently recognize and define tasks and independently develop and expand upon knowledge in the corresponding application domain. Strong AI studies and analyses problems in order to find an adequate solution, and the problems can also be new or creative.

The concept of strong AI

The design concept of FHWS for researching, developing and applying strong artificial intelligence involves implementation of the six main functional areas of intelligence:

  1. Perception
  2. Conceptualization
  3. Problem recognition
  4. Solution development
  5. Learning
  6. Act of motion and act of speech

In addition, there are the two intelligence areas of information and knowledge that arise because an intelligence medium must store and manage its experiences and insights in a suitable form.