Elon Musk's AI company officially apologized on July 12 for the outrageous statements posted by artificial intelligence assistant Grok this week. The reason is believed to stem from a recent software update, which aimed to get Grok to respond as human.
After the upgrade early last week, Grok posted articles on social media platform X, praising the socialist leader Adolf Putin and implying thatJews tend to spread hatred online. Some of the posts were deleted by the X platform hours later, amid the backlash from the online community.
In a post on July 12 on X, xAI wrote: "We would like to express our deep apologies for the terrible behavior that many people have witnessed." The company said it has adjusted the system to prevent similar abuses in the future.
xAI explained that the update requires the chatbot to be set up to answer as a human. This makes Grok vulnerable to extreme views from users, leading to the creation of unethical or controversial responses to keep users interacting.
Grok was introduced by Elon Musk in 2023 as a "gray, straightforward" AI tool with a sharp and un avoided style. However, since its launch, the tool has been constantly embroiled in controversy. In May, Grok was criticized for creating articles with far-right propaganda content that had no basis for the oppression of white people in South Africa, calling it the "white race extinction".
In March, xAI acquired the X platform in a $33 billion deal to integrate social media data into chatbot development. Last week, Elon Musk also announced a new version, Grok 4, unrelated to the July 7 update.
