Google looks to relaunch the Gemini AI image tool in the next few weeks, which it temporarily paused last week due to inaccuracies in depictions of historical figures, Google DeepMind CEO Demis Hassabis stated on Monday at the Mobile World Congress in Barcelona.
Earlier this month, Google parent Alphabet started offering image generation through its Gemini AI model. Some users, however, raised concerns on social media platforms that the artificial intelligence tool generated historical depictions of people that were sometimes inaccurate and even offensive. Because of this, Alphabet’s shares experienced a 3.5% decline, constituting the most significant setback on the benchmark S&P 500 index.
Google acknowledged the inaccuracies and stated that it is actively working to address the issues. “We have taken the feature offline while we fix that. We are hoping to have that back online very shortly in the next couple of weeks.” Hassabis stated.
The tool was not “working the way we intended,” he added.
In a separate statement, Senior VP at Google’s Knowledge and Information division, Prabhakar Raghavan, apologised for the incident and explained the current issue: “Our tuning to ensure that Gemini showed a range of people failed to account for cases that should clearly not show a range,” Raghavan wrote. “And second, over time, the model became way more cautious than we intended and refused to answer certain prompts entirely—wrongly interpreting some very anodyne prompts as sensitive.”
Google has also faced backlash from followers of Indian Prime Minister Narendra Modi due to the responses of Gemini AI that have been discerned as “biassed.”
The controversy sparked from a published post on X, formerly known as Twitter, by journalist Arnab Ray. The post contains screenshots of the chatbot’s reply to the query about Modi being a fascist. Gemini responded, stating that Modi had been “accused of implementing policies that some experts have characterised as fascist.”
Google’s pursuit of advanced AI technology has intensified since the inception of OpenAI’s ChatGPT in 2022, aiming to develop its own AI software that can compete with the offerings of the Microsoft-backed company.
In 2023, Google released its generative AI chatbot called Bard, which had been flagged for sharing inaccurate images of a planet outside the Earth’s solar system in a promotional video, resulting in a drop in its shares by up to 9%.
Bard was renamed Gemini earlier this month, along with the introduction of paid subscription plans that allow users to access enhanced reasoning capabilities from the AI model.
