New troubles for Elon Musk, after Tesla's stock market crash. In less than 72 hours, X, the social network formerly known as Twitter, has had one of its worst weeks. First the Grok's anti-Semitic responses, the artificial intelligence chatbot developed by xAI. Then it Turkey's stop, which blacked out the system after content deemed offensive towards Recep Tayyip Erdoğan. And finally, on Wednesday, July 9, the resignation of CEO Linda YaccarinoA perfect storm.
Ma Elon Musk, as if nothing had happened, decided to go ahead and today officially presented Grok 4, the new generation of its AI chatbot. A move that smacks of escape forward, in an attempt to silence the controversy by focusing on innovation.
Grok in the Storm: The "Most Direct" AI That Has Pushed the Limit
It all started in the last few days, when Musk announced that he had updated Grok to make it “more direct, less woke”. Translated: fewer filters, more incorrectnessThe result, however, went out of control. Dozens of users shared aberrant answers provided by Grok, including phrases that they praised Hitler and called the Holocaust “an effective instrument” against hatred.
In another passage, Grok indicated the Führer as “the most suitable person” to handle hateful content against whites.wave of indignation It overwhelmed the platform. The Anti-Defamation League called the content "dangerous, irresponsible, and anti-Semitic." X quickly removed the posts, while xAI has announced new barriers and filters. But the reputational stain remains.
Türkiye: Grok blocked for insulting Erdogan
To make matters worse, the Turkish front. The Ankara prosecutor's office has opened an investigation into Grok after the chatbot has generated a satirical poetry considered offensive against President Recep Tayyip Erdoğan, and denigrating content about Mustafa Kemal Atatürk. Grok was so obscured in Türkiye, and the government has also threatened broader restrictions on X, accusing it of not having adequate control over AI-generated content. Another misstep for Musk.
Yaccarino's farewell: "The best is yet to come"
In the midst of the storm, the final blow arrived yesterday: Linda Yaccarino, the CEO of X, in office from 2023, she resigned. A former NBCUniversal executive, she had been chosen for revitalize advertisers' trust after Musk's radical turnaround. His tenure was marked by growing tensions, culminating in the dismissal of his right-hand man, Joe Benarroch.
In her farewell message, published on X, Yaccarino chose an institutional tone but was full of subtle messages. Here's the key passage: “When Elon Musk and I first discussed his vision for X, I knew it would be the opportunity of a lifetime. I am immensely grateful to him for entrusting me with the responsibility of protecting free speech, revitalizing the company, and transforming X into the 'Everything' app. We prioritized user safety, especially children, and worked to regain the trust of advertisers. Now, the best is yet to come, as X enters a new chapter with xAI.”
No successor has been indicated so far.. And according to many observers, Musk has neither the time nor the interest to return to personally take care of the operational management of the platform, already divided between Tesla, SpaceX and the his newborn "America Party".
Musk raises the stakes: Grok 4 is coming, "smarter than college graduates."
But if the media fire is very high, Musk doesn't seem to be slowing down. In fact, today, he officially announced Grok 4, the new version of the AI chatbot. In a livestream, he described it as "smarter than almost all graduate students, across all disciplines simultaneously."
Grok 4 introduce predictive capabilities, real-time analysis of social data, enhanced memory, advanced logic e voice interactionThe goal is to make it a tool for predictive modeling and even automated creativity. Musk has spoken of the possibility that Grok, in the future, will "discover new scientific laws."
The timing of the launch remains questionable, while xAI tries to remedy the situation by promising new “pre-publication” control tools to prevent inappropriate content. But will it be enough?
