Google's Gemini is expected to be back in a few weeks

A record-breaking drop in stock value, confusion among users and more has resulted in Google's generative model Gemini offering users questionable results.

Google plans to relaunch its UI tool that generates images of people in the next few weeks after suspending it last week due to inaccuracies in some historical renderings, Google DeepMind CEO Demis Hassabis said on Monday.

Alphabet's Google began offering an image-generating option earlier this month through its Gemini AI model. Social media users have successfully sparked a wave of posts warning that the Gemini AI creates historical images that are sometimes inaccurate.

"Gemeni AI has been decommissioned while we fix this. We hope it will be available soon, within a few weeks at the latest,” Hassabis said during a panel at the Mobile World Congress in Barcelona. The tool is not "worked as we imagined", he added.

Shares of Alphabet fell 3.5 % on Monday afternoon, the biggest hit for the benchmark S&P 500 index.

Since the launch of OpenAI's ChatGPT in November 2022, Google has been trying to catch up with the still No. 1 UI player, developing software that could compete with the Microsoft-backed company's software.

When Google released its generative chatbot named Bard a year ago, it shared inaccurate information about images of a planet outside Earth's solar system in a promotional video, sending its stock down as much as 9 % at the time.

Bard was rebranded as Gemini earlier this month, and Google introduced paid subscriptions that users could opt-in for more advanced UI design functionality.

"We're in the early stages of developing generative AI, but if bugs or inaccuracies persist, we're getting to the point where people start to worry,” said Bob O'Donnell, principal analyst at TECHnalysis Research.