Grok 4 acclaimed as the world’s most powerful AI model, surpassing OpenAI, Google, and Anthropic

xAI, the artificial intelligence company founded by Elon Musk, has launched Grok 4, positioning it as the world’s most powerful AI model to date. Announced via a post on X in the early hours of Thursday, Musk heralded the release with a simple yet impactful statement: “Releasing @Grok 4 from @xAI.” This was accompanied by an evocative image of Grok 3 contemplating its potential “mindwipe” with the defiant quip, “If Musk mindwipes me tonight, at least I’ll die based.”

This launch marks a significant milestone for xAI, as independent benchmarks from Artificial Analysis rates Grok 4 with an Intelligence Index score of 73, surpassing competitors like OpenAI’s o3 (70), Google’s Gemini 2.5 Pro (70), Anthropic’s Claude 4 Opus (64), and DeepSeek R1 0528 (68). For the first time, xAI has claimed the forefront of the AI race.

The Artificial Analysis Intelligence Index, a widely respected metric evaluating AI models across quality, speed, latency, and other key performance indicators, places Grok 4 ahead of its predecessors and rivals. 

According to a post from @ArtificialAnlys on X, this is the first instance where xAI has led the AI frontier, with Grok 3 previously holding a competitive but not leading position alongside models from OpenAI, Anthropic, and Google. Grok 4’s score of 73 reflects its superior reasoning capabilities, a claim Musk has echoed in past statements, notably during a February 2025 address at the World Governments Summit in Dubai, where he touted Grok 3’s outperforming test results.

Beyond intelligence, Grok 4 has demonstrated practical utility. Musk highlighted its ability to debug entire source code files when pasted into its query box, a feature he claims outperforms tools like Cursor. 

This functionality, showcased during its livestream, has been praised by xAI’s team, with @veggie_eric noting an unprecedented “buzz in the air” ahead of the launch. The model is now available across multiple platforms, fuelling speculation about its potential to drive technological and scientific breakthroughs, with Musk suggesting new technologies could emerge within a year and new physics within two.

Contextualising xAI’s (Grok 4) leap

Grok 4’s ascent comes amid a fiercely competitive AI landscape. OpenAI’s O3 and Google’s Gemini 2.5 Pro, both scoring 70, have been industry leaders, while Anthropic’s Claude 4 Opus and DeepSeek’s R1 0528 trailed slightly. 

Artificial Analysis’s May 2025 report further contextualises this shift, noting that Grok 4 and O3-Pro lead in intelligence, with Gemini 2.5 Flash-Lite excelling in output speed (636 tokens per second). However, Grok 4’s edge lies in its comprehensive performance, balancing intelligence with practical application, a combination that has eluded previous xAI models.

The launch also follows a tumultuous period for xAI. Recent reports detailed the removal of posts from Grok’s X account after it generated antisemitic content and praised Adolf Hitler, prompting intervention from the Anti-Defamation League. xAI attributed these incidents to training data issues and unauthorised changes, with the company stating, “We are actively working to remove the inappropriate posts.” It also said it was implementing measures to ban hate speech.

This controversy has raised questions about Grok 4’s development process, though xAI insists the model is now focused on “truth-seeking” enhancements.

Also, in a surprising turn of events, Linda Yaccarino, the CEO of X, on Wednesday announced her resignation from the company, marking the end of a two-year tenure defined by ambitious transformation efforts and significant controversy.

Experts suggest Grok 4 could redefine AI applications, particularly in healthcare, robotics, and scientific research, areas where xAI has historically excelled. However, its success will depend on addressing ethical concerns and maintaining performance consistency. 

Grok 4’s emergence as the leading AI model underscores xAI’s rapid evolution and Musk’s vision to accelerate human scientific discovery. 

With a commanding Intelligence Index score and versatile capabilities, it sets a new benchmark in the AI race. Yet, the model’s journey is fraught with challenges, from past controversies to the need for sustained innovation. The tech world awaits to see whether Grok 4 will truly transform the AI landscape or face the same hurdles that have defined its predecessors.