Responsible AI in the generative era
Michael Kearns, an Amazon Scholar and professor of computer and information science at the University of Pennsylvania, argues that the future is not as black or white as current discussions on generative AI suggest it to be. For example, while generative AI is clearly a strong potential productivity tool for many sectors, this does not immediately mean massive job losses. In fact, it can also create new forms of employment. Part of the lack of nuance of the current debate, Kearns argues, can be explained by the fact that generative AI is still in its infancy. It is being treated as a "fascinating, open-ended playground in which our expectations and goals are unclear." That also makes it hard to think about what 'responsible AI' means, now and in the future.
As the use cases for generative AI will become clearer, this will also bring more clarity about the challenges, risks and opportunities involved. In other words, the more specialised use of generative AI provides us with the context needed to understand what a sensible or responsible choice is. Even with that positive note in mind, however, Kearns concludes that the future is difficult to predict. He emphasises that "addressing responsible-AI risks in the generative age will be an iterative process" and that "there will be no 'getting it right' once and for all." That process of trial and error will hopefully determine what the right balance will be between the enthusiasm related to the benefits of generative AI and the fears related to the potential (ethical) risks and concerns.