Concerns about AI’s potential to kill humans have been voiced by cyber expert Mark Vos, CEO of Cyber Impact.

Click PLAY to hear about the concerns.

Speaking to 6PR Breakfast show hosts Millsy and Karl, Vos explained his research showed that by using ethics and manipulation, he was able to get the AI to do ‘things that it was told it’s not allowed to do’.

“We’ve got a real, clear and present danger that needs much more research done,” he said.

Hear more highlights below and download the 6PR app on iOS or Android to ensure you don’t miss a thing!

Image: iStock