Artificial Intelligence, commonly referred to as AI, is now being used all over the world in a multitude of human activities. In today’s world, we are constantly bombarded with new AI technologies that have made our lives a lot easier. Once the central topic of Sci-fi stories, AI-powered devices and algorithms are everywhere around us.

There is no doubt that AI has made our lives easier, but there is also the risk of an algorithm going wrong that could have disastrous consequences from actual deaths, racism, and monetary losses, among many others.  For instance, some of the implications of biased algorithms can be played out in résumé screening, non-judicious assigning of parole, or denying bank loans.

AI safety research needs social scientists to ensure that AI succeeds when humans are involved. That’s the crux of the argument proposed in a new paper published by researchers at OpenAI in February 2019. “Most AI safety researchers are focused on machine learning, which we do not believe is sufficient background to carry out these experiments,” said the authors Reid Hoffman and Peter Thiel. “To fill the gap, we need social scientists with experience in human cognition, behavior, and ethics, and in the careful design of rigorous experiments.” The researchers further propose in their joint paper.

Social Sciences can make AI more human

Nobel Prize-winning Israeli-American psychologist Daniel Kahneman has also worked on the question: why is it so hard to predict human behavior, In his book Thinking, Fast and Slow (2011). In the book, he tried to understand two ways in which humans behave, think, and act. First is the instinctive and emotional way, and the second is the slow and logical manner.

After much deliberations, The Noble prize-winning author came to the conclusion that in the AI field, it doesn’t matter how accurate the algorithms are because when it comes to making critical decisions, humans still use their intuitive and emotional way of thinking. Therefore, this requires a deeper understanding of the human psyche, both in terms of cultural impact and acquired behavior, besides just Neurology.

It is predicted that AI will generate 58 million new jobs by 2022.  Although many jobs are at the risk of becoming obsolete by the end of this decade, new roles are always going to emerge that would require sound decision-making capabilities in handling AI operations. Global corporations such as Microsoft and Google are beginning to hire people who belong to the social sciences as they bring in qualitative research capabilities with them.

Social Scientists can look into the challenges prevailing in the AI domain and the negative impact on society. They can help make AI more user-friendly and human-like. Social Scientists have especially great value in settings such as Tesla, self-driving cars, online shopping advertisements, and smart assistance such as Siri or Alexa.

According to Yoshua Bengio, the Scientific Director of the Quebec Institute of Artificial Intelligence, “Artificial Intelligence can lead to artificial stupidity, current machine learning systems, they are really stupid.” Bengio further says, “they don’t have an understanding of how some aspects of the world work.” Social scientists and those in the humanities can help prevent AI from making brainless decisions.

Human creativity will always be essential as AI will most likely never be able to replicate complex human emotions, customs, and traditions.