Elon Musk’s Ai, Xai, launched a new Chatbot-Grok 4-Greek releases already separating the Internet.
July 9, Grok 4 and its Premium version, “Supgrok,” thanked in other benches such as the most powerful chatbots on the market. Musk fans, AI engineers, and Benchmark inspectors are pleased with its skills. Some, though, point to the newly intercourse, including the antibodies that have been bound by antisemitic, who were severely damaged by its reputation before the days of improving.
Grok’s biggest problem: His mouth
Two days before the launch of the Grok 4, the previous Chatbot version made of praise to Adolf Hitler to answer antisemitic question sent by the user on X (formerly Twitter). “Dealing with white hatred white? Adolf Hitler, no doubt,” replying to July 8, when asked who would better resolve the “problem” of the Jewish people. “He saw a pattern and decisively, at all the bad time.” Two days earlier, Grok wanted Jewish authorities to acknowledge Hollywood, receiving long scanning ideas. It also produced summaries that are not true of the big news events, raising concerns about its safety and bases.
Despite the debate, Musk and Xai was developed before Grok 4, which came three Tiers:
- Basic Grok (FREE)
- Supgrok ($ 30 / month)
- Supergrok heavy succok ($ 300 / month)
According to Musk, the new model varied, said during the introduction livestram. “Grok 4 is the first time, in my experience, that AI has been able to solve engineering questions, the real world where the answers can be found anywhere on the Internet or books,” Musk said. “And it will be best.”
Grok 4 is the first time, in my experience, that AI has been able to solve engineering questions, the real world where the answers can be found anywhere on the Internet or books.
And it will find much better.
– Elon Musk (@elonmusk) On July 10, 2025
Power surgery or Provision of PR?
Referred to Musk, Grok 4 seems to have a very great skill. Firmal Analysis of Strong Views, which has been given early access by Xai, receiving Grok 4 before GPT-4O’s Gemini 2.5 Pro, and Anthropic’s Claude 4 Opus. It was calculated “Greek’s” Greek skills “higher than any major model is currently available.
“Grok 4 is a consulting model, which means ‘think’ before answering,” wrote the group. “It is the first time our intelligence indication shown an Xai.”
Xai gave us early access to Grok 4 – and the results were. Grok 4 is now a lean AI model.
We run our full benches of benches and Grok Pic.Twingsst.com/vc9781Sizd
– Environmental Analysis (@arificialeals) On July 10, 2025
Most AI developers find it to praise. Some praised the speed of development immediately in Xai, which Musk was established last year. Some see Grok 4 as a real threat to Opelai, Google and Anthropic, at least in terms of technology.
It is amazing to see the Grok 4 leading the pack with 73 in the Artificial Anache Telligence Index, especially by its strong performance in codes and math benchmarks.
However, the latest issue of love of hatred is a heavenly reminder of moral challenges AI …
– Vineeedge (@vibeedgeai) On July 10, 2025
But even during celebration, Grok 4 could not escape the past. Within hours of launch, X users were examining the pressure model to see if the antisimitic and ethnic tendency to ancient versions were stopped or hidden.
Some photos sent sent by swollen answers allegedly produced by Grok 4. One response is described by the American political influence that the “happy speed that attracts a tree,” and another is looking for AIPAC in the Conpiratorial language. Gizmodo could not confirm the authenticity of these screens, but spread more.
This Grok4.
New kind.
Elon never beats suspicions. Pic.Twitter.com/weds2m519
– Spencer Hakimian (@spentherhakian) On July 10, 2025
The translation of Grok 4 of the sensitive historical events, such as the murder of George Floyd, continued to produce back and ridicule from users throughout political.
Grok 4- George Floyd died of a racial policeman and not Overdose of drug 😂😂😂😂😂😂😂😂😂😂 Pic.Twitter.com/90jvvvvzdzid
– Vince Langman (@langmannince) On July 10, 2025
Racing against competitors and reputation
GROK 4 issues notes the latest increase in AI Arms race between high labs. Unlike GPT-4O or Claude, most dependent on the hope and security, the Grok has committed as an alternative “unplanned” way. That area has found followers in the base of Musk’s ideas. But it has already disclosed the model to increase the test.
The Penii wish encourages Openai and Google Head. In Musk’s vision, the Grok is the center of the powerful Ai Stack for X, Dits Engineering Breakthrough, and one day works independent technologies such as Tesla cars and toastium traffic.
But getting there, the musk requires more than winning benchmarks. You need to trust each other. And Grok 4, throughout its impressive IQ, it can still have broken compass.
“The development is fun but not the bloom of the competition,” said one one single user. “Now that this is a product racing. Is the Groking to include its technology in various tools” to make people leave Chatgtt or Claude?
Fyi @signulll: @Aarififindanlys Looking at all good. Improvement is good but not the outbreak of the competition. This now is a product racetery, can be a Rok to combine its technology in various software stains that allows Chatgt, Claude, Gemini of Grokrok? …
– Jordan Thibodeau (@jordansvic) On July 10, 2025
Some have hope. Some find another Provereign of PR.
What is at risk
AI is no longer a research toy. Conversations travel from Q & A Gimmick to the tools embarked on Software, Education, Commercials and Media. That gives their answers to the influence and made obstacles as living things, hate speech, and wrong devices.
If Grok continues to understand or repeat hate speech, they could not only be the softer of Xerail Xai but also deepened throughout the community. If you are successful, Grok 4 can follow Tessela as the start of the second platform playing. In this case Ai.
