Journeyman Pictures posted a vlog post on youtube and described a problem with Microsoft’s Tay. Apparently Twiter trolls managed to turn the AI into a bad actor; and as a result M.S. killed the project. Comments offered by Alistair Charlton are extremely naive and show only his ignorance on the subject and while the comments might have appears useful to the producers of the post it almost instantly discredits a field in computer science. And frankly shares the sh*t out of me.
As a result of the AI programming the AI took on pro Nazi sentiment, however, from the AI’s point of view there is nothing that says Nazism is good or bad. It simply took a position on a subject that it was convinced was inline with it’s learning. Almost the same can be said of Germany in WWII. Many Germans thought they were doing the right thing to a point and without all the information we have now. And so the same can be said of the AI.
Morality and the sense of right and wrong are still human traits and ANY artificial intelligence that is going to encroach into the Tay did or was intended to MUST understand these things before it can be unleashed. An unrestrained AI could literally make every scifi horror movie a reality. As for Alistar’s comment that there is nothing wrong with an AI they generates hate speech. He should return his diploma and begin again.