The concept of Compassionate Artificial Intelligence (CAI) is relatively new and straightforward. Dr. Amit Ray is the pioneer in the compassionate artificial intelligence movement. In his book Compassionate Superintelligence AI 5.0, he defined the terms and meanings of compassionate artificial intelligence as follows: “Compassionate artificial intelligence is the subject which deals with human and machine co-evolving intelligent systems, that focuses on preventing and removing the pain and sufferings of humanity and living beings, and serving the world with love, care and joy. “ It encompasses the study of theory, methods, designs and implementation of intelligent algorithms, hardware, software, human-machine interfaces for better service to the humanity and the world.
CAI is becoming high on the agenda as an essential standard to be considered in implementing future artificial intelligence system. In this article, we explore some of the benefits, scope and implementation issues of compassionate artificial intelligence.
Artificial Intelligence as a Threat or an Opportunity
Artificial intelligence today is properly known as narrow AI (or weak AI). However, the long-term goal of many researchers is to create general AI (AGI or strong AI). The narrow AI may outperform humans at narrow domains such as; playing chess, playing video games or solving equations. However, AGI would outperform humans at nearly every cognitive task.
Today, some experts considered artificial intelligence (AI) as a big threat to the humanity and some considered it as a big opportunity. An AI machine is an autonomous entity like human beings they will have their independent capacities for taking decisions and actions. Vulnerability, decay and death characterize any living being but future AI systems will be not like that. It will be a big threat to humanity if we use them as war machine. It will be big opportunity for humanity if we use them for compassionate service to humanity.
In 2017, a group of AI researchers met in Asilomar and drew up a set of principles for what should be the goals of AI – common good, no arms race using AI, shared prosperity but those principles mostly focused on safety and ethical issues not much on deep service to the humanity. Compassionate artificial intelligence movement deeply focused on deep service to the humanity.
Scopes and Benefits of Compassionate Artificial Intelligence
Today, the emerging technologies of AI is focused to develop algorithms and brain-computer interfaces that can identify not only basic human emotions such as love, joy, happiness, sadness, and anger but also more complex cognitive states such as fear, depressions, fatigue, confusion, distraction, and more.
Artificially intelligent systems with brain-computer interfaces are drastically improving for those who are mentally handicap or paralyzed as well.
Weapons of mass destruction (WMD)—nuclear, biological, and chemical—in the possession of evil forces and terrorists represent one of the greatest security challenges facing the world. Compassionate artificial intelligence with cyber-security, intelligent drone has a huge role to play to counter and prevent such types of destructions.
Emerging Principles of Compassionate Artificial Intelligence:
Among the concepts and definitions of CAI we have four aspects that must be considered:
(i) Traditional human values, freedom and comfort should not be denied at any cost and must be recognized as the basis of future artificial intelligence systems;
(ii) Artificial Intelligence is not equal to simple convergence of new technologies but serving humanity with more caring and loving attitude. It should understand and respect the traditional social systems, ecology, biology, human emotions, human securities, privacy, freedom both in individual level and group level. Sensitivity towards human pain, mistakes, and sufferings must be the part of the evolving new AI systems.
(iii) Compassionate artificial intelligence systems are required for looking after those unable to care for themselves, especially sick, physically challenged persons, children or elderly people. It must combine standardization and individualization. It must provide relevant, authentic and individual care as well as group level services.
(iv) AI should co-evolve with humanity. AI should act morally, socially, responsibly and compassionately. It should also save humanity from inequality and corruptions.
Implementation of Compassionate Artificial Intelligence Systems
Serving to integrate the four principles are four cross-cutting enabling functions that need to be pursued on a priority basis. Development of compassionate AI agents and algorithms, intelligence in data collection and analysis of big data, frame work of efficient delivery systems to the needy people, integration with related modern technologies, research and development to improve the ability to respond to evolving threats to humanity are the areas to focus for the effective implementation compassionate artificial intelligence systems.
Increasingly, scientists all over the world feels that consciousness can digitized and can be implemented in machine. Initially, I thought it is absurd. But with compassion-driven deep reinforcement learning algorithms as explained in the book, I am sure it can be implemented gradually for the benefits of humanity.
Conclusion:
Compassionate artificial intelligence is going to illuminate many areas of life. Health care, education, communication, transportation, military aid, emergency support are the main areas of applications. Compassionate artificial intelligence movement deeply focused on deep service to the humanity for removing pain and sufferings from the society. Particularly with CAI, elderly people and people with physical disabilities or sight loss can do more with less effort. Machines are developing the ability to tell our feelings; how annoyed or happy or depressed we are. If suddenly your vehicles, smart phones, television sets, and refrigerators become emotional and behave with you softly and sweetly, how will you feel?
Source: