The alarming exchange unfolded when Reddy asked the chatbot for help with a school assignment exploring challenges faced by older adults. Instead of providing constructive assistance, the chatbot issued a series of disturbing and hostile statements,

"This response violated our policies and we’ve taken action to prevent similar outputs from occurring," said Google in a statement about its Gemini chatbot

AI chatbots put millions of words together for users, but their offerings are usually useful, amusing, or harmless. This week, Google’s Gemini had some scary stuff to say.

The student from Michigan, US, was having a conversation with the chatbot about a topic of their homework, when it threatened the user.

In an online conversation about aging adults, Google's Gemini AI chatbot responded with a threatening message, telling the user to "please die."

Google's glitchy Gemini chatbot is back at it again, folks — and this time, it's going for the jugular. In a now-viral exchange that's backed up by exported chat logs, a seemingly fed-up Gemini begs a user to "please die" after they repeatedly asked the chatbot to solve their homework for them.

Google has said it's chatbot it designed to filter out potentially harmful responses but this is not the first time the company has come under criticism for it's AI chatbot