In a recent development, Google has decided to pause its AI portrait tool following complaints about historical inaccuracies. The Alphabet-owned company has been in a race to develop AI software that can rival the offerings from OpenAI and Microsoft.

The tool, known as Gemini, was introduced earlier this month as part of Google’s image generation capabilities. However, users on social media have pointed out instances where the model produces historical images that deviate from accuracy. Notably, some depictions portrayed the U.S. Founding Fathers as women or Nazi soldiers as people of color.

In response to these concerns, Google acknowledged the inaccuracies in historical image generation associated with Gemini. The company is actively working to address these issues.

This move comes in the wake of OpenAI’s successful launch of ChatGPT in November 2022. Google aims to keep pace with the advancements made by its competitors, striving to offer comparable AI software solutions.

Interestingly, Google faced a similar situation in the past when its generative AI chatbot, originally named Bard, provided incorrect information about a planet beyond our solar system in a promotional video. Now rebranded as Gemini, the chatbot is part of Google’s efforts to enhance its AI offerings, including the recent introduction of paid subscription plans.

As Google continues its pursuit of cutting-edge AI technology, it remains committed to refining its tools and ensuring historical accuracy in image generation.

11 COMMENTS

  1. They have to redo the fine tuning and that takes time. They probably need to build a new fine tuning training set. That’s take time too. Hope it comes out better and more reliable.