Google is urgently working to address issues with its recently launched AI-driven image creation tool amid allegations of excessive correction to mitigate the risk of racial bias.
Customers reported that the Gemini bot from the company provided images representing diverse genders and ethnicities, even in cases where such representations were historically inaccurate.
An inquiry for images of America’s founding fathers yielded results featuring women and individuals of diverse racial backgrounds.
The company acknowledged that its tool was not meeting expectations.
“Although Gemini’s AI image generation produces a diverse array of people, which is generally beneficial for its global users, it’s not hitting the target in this instance,” remarked Jack Krawczyk, senior director for Gemini Experiences, during Wednesday’s announcement.
“We’re actively striving to enhance these types of representations without delay,” he further explained.
Following its release, Google announced it would temporarily suspend the tool’s capacity to generate images of individuals while it addressed the issue. This incident underscores the recurring challenge AI faces in addressing real-world concerns regarding diversity.
A notable instance dates back almost a decade when Google faced backlash and issued an apology after its photo app mislabeled a picture of a black couple as “gorillas.” Similarly, OpenAI, a competitor in the AI field, encountered criticism for potentially perpetuating stereotypes. Users discovered that its Dall-E image generator tended to produce results predominantly featuring white men when prompted with queries like “chief executive.”
Amidst mounting pressure to demonstrate its advancements in AI, Google launched the latest version of Gemini last week. This bot generates images in response to written prompts. However, it quickly faced criticism from skeptics who accused Google of training the bot with overly conscious, or “woke,” parameters.
Computer scientist Debarghya Das remarked that it’s embarrassingly difficult to prompt Google Gemini to recognize the existence of white people. Author and humorist Frank J Fleming, known for his contributions to outlets like the right-leaning PJ Media, expressed his disbelief when he received results requesting an image of a Viking.
These assertions gained traction within conservative circles in the United States, adding to the existing criticism against major tech platforms for perceived liberal biases.
Mr. Krawczyk emphasized the company’s commitment to addressing representation and bias, aiming for search results that reflect its diverse global user base.
He stated on X, formerly Twitter, where users were sharing questionable results, that historical contexts carry nuanced complexities and that they are dedicated to refining their algorithms to better accommodate these nuances.
“This process is part of our ongoing alignment efforts, involving iterative adjustments based on feedback. We appreciate your input and encourage you to continue sharing it.”
Read More: Google halts Gemini AI image generator due to inaccuracies in historical images it created
Disclaimer:
This content is AI-generated using IFTTT AI Content Creator. While we strive for accuracy, it’s a tool for rapid updates. We’re committed to filtering information, not reproducing or endorsing misinformation. – Jomotoday for more information visit privacy policy
Leave a Comment