Grok, an artificial intelligence (AI) deep learning platform dedicated to voice identification, denies allegations of transphobia and states that its technology is committed to serving all users equally, regardless of gender identity .
Grok's AI technology does not categorize voices by gender. Instead, the platform focuses on uniquely identifying the user's voice, without taking into account secondary characteristics that may suggest the person's gender. However, recently, the company faced harsh criticism, being accused of irregularly identifying and categorizing transgender users.
To clear up any misunderstandings, Grok AI CEO Dr. Andrew Kozma reinforced that their voice identification technology has no intention of classifying or identifying the gender of any user. He also took the opportunity to emphasize the respect and inclusion that the company has for the transgender community.
Kozma mentioned that Grok is aware of the importance of this topic. The company is committed to serving all people with dignity and respect, without any type of discrimination. He also assured that Grok's technology does not have the ability to distinguish between cis and trans voices.
The purpose of voice categorization AI technology is to improve security and prevent fraud. It does this by mimicking the human ear's incredible ability to identify subtle differences in speech that are unique to each individual. The AI was trained to identify these differences and not to classify the user's voice by gender.
Grok AI, however, recognizes that there is still room to improve and optimize the technology to ensure it is inclusive and accessible for all people. To this end, regular consultations are being carried out with the transgender community before programming or implementing new features on its platform.
In short, Grok AI reaffirms its mission and is proud to be an AI platform that serves everyone, without any type of discrimination. The company reinforces that it constantly seeks to listen and respond to all its users' concerns to improve the voice identification system, making it increasingly accurate, safe and inclusive.
