Artificial Emotional Intelligence Testing for AI Avatars

There has been significant dialogue over the last few years regarding the utilization perils of artificial intelligence in virtually everything we touch. From our phones to our cars, and everything in between, artificial intelligence has become an integral part of our existence. Nikola Tesla described the dangers of unrestricted AI when he expressed that a “borrowed mind” built from human minds may not be able to contain their inherent flaws or biases”. Prominent figures, like Elon Musk and Stephen Hawking, have also warned about the potential for machines to take over and wreak havoc in the lives and existence of humans. Hollywood has made untold billions of dollars painting doom-and-gloom scenarios about artificial intelligence and robots within our society today and in the future. But what is the true reality? We continually push to create increasingly intelligent systems/machines that attempt to learn, think, and reason like humans. Most attempts fail miserably. Given the recent “issues” with AI avatar chatbots like ChatGPT ^® we enter a brave new world of interactions with AI-like entities. These AI avatar entities have no cognitive functions and do not understand human emotion, only human language and primarily only English, currently ignoring the other ~3000 written and ~10000 spoken languages. Here we discuss the need and a framework for understanding human emotions in order to create better human-AI communications, and we present an artificial emotional intelligence test that can be used to determine an AI avatar's ability to, and improve the capacity to, functionally communicate with people and between other AI entities.