Dangers and malfunctioning are two significant risks. Last year a chatbot named “Tay” used by Microsoft for Twitter posted Nazi statements. Another shocking example of the failure of AI isUber unauthorized/unapproved test of its self-driving cars which ran over 6 red lights, and one of the car ran a red light at a busy pedestrian crossing, this sounds so dangerous and unsafe for other people on the road.
Talking about Facebook’s photo tagging, although it seems a great work of AI this is not error free. Zhang (2016) states “When Brooklyn-native Jacky Alcine logged onto Google Photos on Sunday evening, he was shocked to find an album titled “Gorillas,” in which the facial recognition software categorized him and his friend as primates.” Later Google had to apologize for this.
There are several malfunctioning incidents. Depending on machines totally can never assure us to be error free or danger free.
References:
Zhang, M. (2016) Google Photos Tags Two African-Americans As Gorillas Through Facial Recognition Software. Available at: https://www.forbes.com/sites/mzhang/2015/07/01/google-photos-tags-two-african-americans-as-gorillas-through-facial-recognition-software/#3d10e5ca713d (Accessed: 19 June 2017)
