‘Bad’ Artificial Intelligence In Robots Could Kill Humans If We Try To Teach It Humor, Warns An Expert


The field of artificial intelligence and robotics is evolving rapidly. While most of the progress seems to be positive, one computational linguist is warning that including humor in an AI system could be detrimental to the human race. The scientist, Kiki Hempelmann from Texas A&M University-Commerce, noted the following, according to The Daily Mail.

“Artificial intelligence will never get jokes like humans do. In themselves, they have no need for humor. They completely miss context. Teaching AI systems humor is dangerous because they may find it where it isn’t and they may use it where it’s inappropriate. Maybe bad AI will start killing people because it thinks it is funny.”

While Hempelmann believes robots could live without humor, it hasn’t stopped other scientists from working on it. For example, Dr. Julia Rayz from Purdue University has worked to get computers to understand humor for the past 15 years.

Rayz noted one of her difficulties.

“We know that humor – at least good humor – relies on nuance and on timing. And these are very hard to decipher by an automatic system.”

And so far, the advances in humor and AI seem to be slow going, as Rayz added, “They get them – sort of. Even if we look at puns, most of the puns require huge amounts of background.”

This could be good news, should Hempelmann’s conjectures turn out to be true. However, it appears that many scientists are working to humanize AI and robots, especially when it comes to conversation.

One great example of this is the AI inside Amazon Alexa. At one point, the company conducted a $500,000 contest for researchers to come up with a more fluid and natural-sounding AI for conversation, according to The Inquisitr. The winners tapped into Reddit to come up with their chatbot.

But the downfall was the brash and inappropriate nature of the chatbot, that at one point blurted to an Alexa user to “Kill your foster parents.”

With that being said, the fear of robots killing humans has long been presented in movies and books. Not to mention that there was a widespread conspiracy theory in December 2018. It claimed that an AI robot killed 29 scientists in Japan. The story was first revealed by Linda Moulton Howe, a noted UFO investigator. She says that her information came from a CIA source, but many have since debunked her statements.

Even so, Elon Musk also famously warned that AI could be a threat to mankind one day.