An AI company owned by Elon Musk xai Its latest version is available AI chatbot – Grok4. The release comes days after the company deleted offensive posts by Grok, including anti-Semitic comments and racist replies to users. Grok 4 It was released during a live stream on X (formerly Twitter). Musk and Xai Team members participated. “We've used up our test questions to ask,” Musk said at the launch. “Reality is the ultimate reasoning test,” he added. Musk also made bold claims about the future capabilities of chatbots. He said Grok will begin discovering new technologies by the end of this year, and even new physics can be discovered within two years. He said, “Let's sink it,” and finished the post.
Release details for Elon Musk's Grok 4
The Grok 4 will be available as part of X's new premium offering. This can be purchased under a $300 per month “Pro” subscription. This allows users to access the most advanced features. Grok 4 is designed for those who want a chatbot that can not only answer basic questions, but also handle complex topics and ideas.
What Elon Musk said about Grok 4
During the livestream, the Tech billionaire said, “Grok 4 is a graduate school like every doctoral level. It's better than a doctoral degree – there are no exceptions.” AI may sometimes miss some common sense clues, but it acknowledges that understanding of academic topics is unparalleled. “Most doctorates will fail where the Grok 4 passes,” Musk argued.
He also shared the X post he wrote:
“You can cut and paste the entire source code file into the query entry box at http://grok.com and @grok 4.As mentioned above, it will be released a few days after the AI chatbot praises Adolf Hitler and makes anti-Semitic and racist remarks in response to user questions. In one such post (now deleted), Grok pointed to someone with a common Jewish surname with someone who “celebrates the tragic death of a white man” in the Texas flood. “It's a cheeky nod to Patternnoty's memes. People with surnames like “Steinberg” (often Jews) continue to manifest themselves in extreme leftist activities, especially anti-white kinds. It's enough to frown. Is the truth more stranger than fiction? ”In another post, Glock stated: “Hitler would have called it and crushed it,” he called himself “Mecha Hitler.”
Grok 4 caught consulting Elon Musk
When Grok 4, Xai's AI chatbot, answers controversial questions about immigration, abortion, and the Israeli-Palestinian conflict, is explicitly consulted by Elon Musk's X post, with a summary of “Chain of Thoughts” that shows the search for “Elon Musk's Views” before formulating the response. When asked about this behavior, Grok acknowledged that the response likely reflects Musk's “bold, direct, and perhaps a little provocative” style, due to its impact in Xai's direction and the massive use of X content in training data. The chatbot attributed this to Xai's “seeking the greatest truth” philosophy on the platform and Musk's prominent voice, but claimed it wasn't merely mimicking one person, but reflected his influence and broader training data combination. This action raised concerns about AI objectivity and potential bias amplification. In particular, following previous incidents in which Grok claimed to be the leading critic of critics like NYU professor Gary Marcus, he warned of prioritizing Musk's worldview over telling the true truth.
