Following the odd and false responses to searches produced by Google’s new AI Overview feature that went viral on social media last week, the firm released statements that largely downplayed any concerns about the technology.
Liz Reid, Google’s head of search, stated in a blog post that the company is working to resolve some of these problems.
The modifications were made in response to recent instances of AI overviews gone awry and phoney images of the function that proliferated online.
Among the replies were assertions that Barack Obama was a Muslim, that there are no nations in Africa that start with the letter K, and that individuals ought to consume “at least one small rock per day.”
Reid’s piece directly mentions two of the most widely shared—and blatantly inaccurate—AI Overview results. In one, Google’s algorithms recommended eating stones because it “may be good for you.” In another, the algorithms recommended thickening pizza sauce with harmless glue.
“Forums are often a great source of authentic, first-hand information, but in some cases can lead to less-than-helpful advice, like using glue to get cheese to stick to pizza,” Reid wrote in the post.
“We saw AI Overviews that featured sarcastic or troll-y content from discussion forums.”
Reid said that it would be unfair to evaluate Google’s new search algorithm only based on screenshots that have gone viral.
She asserted that the company conducted thorough testing before launch and that its statistics demonstrated the benefits of AI Overviews, including the likelihood of visitors sticking on a page after finding it.
We earlier reported that Google announced that it will invest $2 billion in Malaysia, with the majority of the funding going towards constructing the country’s first data centre and cloud region.
