People cannot believe how dangerously wrong new Google feature is

People have been calling out the new feature Google added to their search engine in the US after it started providing them with some very incorrect and at times dangerous answers.

Google said their new AI Overview tool would 'do the work for you' in making information on the internet easier to find, but they've been tripped up by some of the results it's provided.

What AI Overview is supposed to do is produce a summary of the search results so you don't have to sift through a bunch of entries trying to find the answer.

You know how it goes, run into a problem or need a quick answer and Google will provide you the answer if it exists on the internet.

Only, it's up to you to go through all of the entries and dig around for the answer, which is what this AI is supposed to remedy.

Some people say they've found problems with Google's AI, the company says they're working on it. (Jakub Porzycki/NurPhoto via Getty Images)

Say, for example, you've got a tech issue with your computer and the solution doesn't seem to be obvious, but you know that somewhere on your computer is a way to fix it.

Chances are someone on the internet has discussed this very issue before and there'll be an answer buried on page six of some forum from 2011, which if you hop on Google and search it might not be close to the top answer.

In theory, AI Overview would find the answer you seek and stick it at the top of Google - but in practice, it's been quite different for some.

Users have reported that AI Overview has told them to smoke between two and three cigarettes a day while pregnant, to use glue to keep cheese on a pizza and that it's healthy to eat 'at least one small rock a day', which you definitely shouldn't do as it's monstrously dangerous.

A number of people have also found that Google's AI said former US president Barack Obama was Muslim, which is a conspiracy theory cooked up by his opponents.

Tech companies ought to know by now that for all they can do, AI is also very vulnerable to being fooled.

Some users have been posting the dangerous results AI provided them. (X/@floriyrobin)

You can see the problem here, and according to CNN, the tech company has now started taking down some AI responses which they say violate their rules, and Google also made a statement on the matter.

A Google spokesperson told LADbible: "The vast majority of AI Overviews provide high quality information, with links to dig deeper on the web.

"Many of the examples we’ve seen have been uncommon queries, and we’ve also seen examples that were doctored or that we couldn’t reproduce.

"We conducted extensive testing before launching this new experience, and as with other features we've launched in Search, we appreciate the feedback.

"We're taking swift action where appropriate under our content policies, and using these examples to develop broader improvements to our systems, some of which have already started to roll out."