New York: Elon Musk’s startup XAI apologized on Saturday by offensive publications published by its artificial intelligence assistant Grok this week, blaming them for a software update aimed at making it work more as a human.
After Tuesday’s update, Grok praised the Nazi leader Adolf Hitler in the publications on the social media platform X, and suggested that people with Jewish last names were more likely to spread online hatred.
X eliminated some of those publications several hours later, in the midst of the growing outrage.
“We apologized deeply due to the horrible behavior that many experienced,” the company published on Saturday X, adding that it had modified the system “to avoid more abuse.”
The company said the change occurred after the chatbot was requested that “respond to publication as a human”, as well as “tell me how it is and you are not afraid to offend people who are politically correct.”
As a result, Grok became susceptible to the “extremist opinions” of the users, who made it produce “responses that contain unconverted or controversial opinions to involve the user.”
Grok, which Musk promised that it would be a true “avant -garde” after its launch in 2023, has been involved in controversy.
In March, XAI acquired X in a $ 33 billion agreement that allowed the company to integrate the platform data resources with the development of the chatbot.
In May, Grok turned on the controversy by generating publications with right -wing propaganda without support on the alleged oppression of white South Africans that he called “white genocide.”
On Wednesday, Musk presented a new version of the wizard, Grok 4, which was not related to the July 7 update.
