[Tyler]

Here's How We Prevent The Next Racist Chatbot
24 March 2016, 12:00 am

                    It took less than 24 hours and 90,000 tweets for Tay, Microsoft's A.I. chatbot, to start generating racist, genocidal replies on Twitter.

Popular ScienceLink

Source: AI in the News

To visit any links mentioned please view the original article, the link is at the top of this post.

[Don Patrick]

Don't forget to tell it to learn from the story's protagonist then and not from the bad guys' behaviour. How do we tell the bad guys apart? Because they do bad things. How do we know that these things are bad? Because the bad guys do them. And right back where we started.

Meanwhile a basic sentiment analysis database will shower you with thousands of words marked as bad. How about use that. Or just don't build parrot bots.

[DemonRaven]

Don't forget to tell it to learn from the story's protagonist then and not from the bad guys' behaviour. How do we tell the bad guys apart? Because they do bad things. How do we know that these things are bad? Because the bad guys do them. And right back where we started.

Meanwhile a basic sentiment analysis database will shower you with thousands of words marked as bad. How about use that. Or just don't build parrot bots.

Yes that is about it as you need to train them. No matter what data base you use eventually you still have to go in a add stuff because language is not static.