Information Systems Integration – Messina

Training AI to Benefit Society with Core Values

AI photo

As Artificial Intelligence (AI) is growing in it’s capabilities, businesses are aiming to train it to benefit society as best as possible. With AI growing we must consider how it can impact other people’s lives not merely as a tool but as a phenomenon that will be converged into everyday life. To do so businesses have experimented by trying to teach AI core business values. These values include a sense of responsibility in work, respecting customers, fairness, and transparency.

Do you believe that AI is capable of learning such core ethical values? Where do you think the limits will be if there are any? With technology growing and AI being a part of that, is it appropriate to try to hone and teach it core values or do you think we should abandon such technology for whatever negative effects it may bring?

To learn more about this, click the link provided below.

Source: https://www.forbes.com/sites/steveolenski/2018/03/12/the-5-technology-trends-that-may-disrupt-business-over-the-next-3-years/#3318ca494b18

3 Responses to Training AI to Benefit Society with Core Values

  • I think that AI having ethical values is kind of crazy to think about. We think of computers as machines that don’t really have feelings or emotions, but maybe they could develop such tendencies. It would certainly change the future of computing and our lives.

  • This is an interesting read. I do believe it’s possible for AI to simulate these core values and emotions through more complex disciplines of AI such as machine learning. Using enormous data sets, machine learning could analyze human responses/reactions to many situations then generate and simulate a value system accordingly. In theory, machines can simulate these values and emotions; however, I don’t believe it’s possible for machines to develop true, real values and emotions.

  • With any technology there are benefits and drawbacks. AI is no different. The AI has learned the gender and race bias’ that still exists in society today, and willl reinforce that bias unless we teach it otherwise.

    So for example, in a test of Google’s AI, the AI associated black people with gorillas, and in its photo tagging capability suggested a gorilla tagging instead of a person. This also works for gender, Google also has gender bias. A researcher asked the AI to complete a sentence “Man is to computer programmer as woman is to X,” it replied, “homemaker.”

    It points out that these bias’ still exist, because the technology is learning from us. However this may be a positive, it proves we still discriminate on a social level, and may force people to rethink their bias.

    https://www.wired.com/story/machines-taught-by-photos-learn-a-sexist-view-of-women/

Leave a Reply to Jacqueline Henry Cancel reply

Your email address will not be published. Required fields are marked *